IT, in its current form, relies entirely on networks. From office LANs to transatlantic fibre-optic cables, to WiFi and cellular connections, no business can really carry out its day-to-day tasks without using at least one network.

Yet, with all the attention focused on AI and its vociferous appetite for GPUs, CPUs, and even storage, the network risks being overlooked.

Without both investment and new ways of engineering the network, connectivity could derail progress in digital transformation, the user and customer experience, data science, and, of course, AI.

Recent research from vendor Cisco found that three in four enterprises expect to hit the limits of their campus and branch networking capacity within two years, and they are seeing network traffic double year on year. Already, some enterprises say they are holding back on scaling up AI investments because of network constraints.

Unlocking the problem

“You have to look at where bottlenecks are,” Paul Stringfellow, CTO at IT consultants Gardner Systems, told ITPro.

“This is less an enterprise network issue, but more a backend issue. So, looking at how you ensure high capacity [connections] between CPUs, GPUs, and back-end storage.”

As Stringfellow points out, there have been advances in high-speed connectivity, including faster Ethernet speeds and RoCE (RDMA over Converged Ethernet), which allows remote direct memory access over Ethernet technology.

These, though, along with Infiniband and 400Gbps and even 800Gbps Ethernet, are mostly targeting data centres and cloud providers.

For client devices, developments in WiFi should improve both connectivity for users and for IoT devices. Technologies such as WiFi 6 and WiFi 7 – and in the future, WiFi 8 - look to increase bandwidth and provide reliable connections, even as more users or devices connect to the WLAN.

“WiFi also needs to be included in this conversation, so many users now use WiFi as the primary office connection. The idea of cabled connections is foreign in a lot of places,” says Stringfellow.

The move to cloud and SaaS-based applications, as well as widespread home working, means that much of an enterprise’s data traffic has shifted to WiFi, with users never reaching for an Ethernet cable. Effective WiFi has become essential, with poor connections directly affecting productivity.

“Enterprise Wi-Fi has evolved significantly and now carries far more than simple user connectivity,” Phil Huang, business development and field application manager for the UK and Ireland at vendor D-Link, told ITPro.

“Modern Wi-Fi networks routinely support collaboration platforms, voice and video applications, IoT devices, and a wide range of cloud-based services,” he says. “In that sense, Wi-Fi has become a strategic access layer rather than just a convenience technology.”

But, he says, there are still significant differences between wired and WiFi networks. And improvements in WiFi connection speeds are pushing bottlenecks further into enterprises’ core networks. This, in turn, is driving investments in the wired LAN too.

“Newer Wi-Fi 6E and Wi-Fi 7 access points are capable of exceeding 1Gbps throughput, meaning a standard Gigabit uplink can quickly become the bottleneck,” says Huang.

As a result, businesses are moving to both 10Gbps connections and 2.5GbE. This is moving closer to “standard” 1Gbps connections in cost. “The premium over traditional Gigabit Ethernet has reduced dramatically, making 2.5GbE an easy investment for organizations refreshing their infrastructure,” he says.

Installing 2.5GbE, and even 10GbE, is simpler than fibre-optic or even higher-speed copper LANs. But it still works well for most office deployments, at distances under 100 metres.

Campus to cloud

The bigger challenges lie both in connectivity from businesses to data centers and the cloud, and within the data center itself.

Here, growing data volumes and the need to reduce latency are prompting CIOs to look at how their networks are designed, as well as the hardware they run on.

“As SaaS and AI adoption grows, hybrid work reshapes access patterns and digital security follows users beyond traditional perimeters,” Nauman Raja, a director analyst at Gartner, told ITPro.

“Enterprise networks are no longer built around the data centre. Instead, they now gravitate around the internet.”

AI too is playing its part: expensive GPUs are a waste of money if the network cannot keep up. Network architects need to meet the high-bandwidth demands of uploading data for model training, usually to the cloud, and the low-latency needs of inference.

Organizations also need to consider how they connect their existing enterprise applications to AI models. Conventional networks are rarely built with these demands in mind.

“I don't think we're having a bandwidth problem. We're mostly having a design or infrastructure problem,” Wim Remes, a technology consultant who has also run operations for a managed security service provider (MSSP), told ITPro.

“Compute and data are not always in the same place. Either we bring data to the compute or we bring the compute to the data. The paths we create on the network need to be the shortest.”

“The top-level issue, to me, is that we’ve ignored ‘the network’ for about a decade and a half,” he adds.

“Does a full-stack developer know the basics of networking? The clear answer is no. The stack in ‘full stack’ stops at the web server at best. At the same time, many a CIO also doesn't really care about the network anymore.”

The challenge is to bring together new and emerging hardware technologies, such as 400 or 800Gb Ethernet and RoCE, with a network architecture giving the performance data-heavy, latency-intolerant applications need.

“Bad network design, such as routing internet-bound traffic through centralised data centres for software as a service (SaaS) application, or for end-user collaboration, such as video conferencing, will result in poor user experience and inconsistent access for both remote and on-site users,” Gartner’s Raja warns.

“As demand fluctuates with remote working patterns, network architectures must allow dynamic bandwidth allocations,” Raja adds.

“In many cases, smart traffic management and automated policies can reserve sufficient capacity for business-critical applications even during peak usage.”

That, though, means thinking less about the network as hardware, but as an operational issue across the business. After all, no one wants colleagues, or worse still, customers, locked out of applications because someone “tokenmaxxing” is blocking the LAN.

What is HPE? For just over a decade, since its split with what's now HP Inc, the answer has been clear: HPE is an enterprise hardware company specializing in servers and storage.

The additions of Cray and SGI to its portfolio bolstered this position: While they primarily opened the door to HPC and supercomputing for HPE, fundamentally there's a lot of crossover between typical data centers and these behemoths.

Even the name – Hewlett Packard Enterprise – invokes a sense of a company with a long history of making specialist kit for organizations' data centers or server rooms.

At HPE Discover 2026, though, you would be forgiven for thinking HPE's primary competitor wasn't Dell or IBM but Cisco.

Both of the first day keynotes were chock full of networking. I'm not just talking about announcements, of which there were some, I'm talking about the overall vibe as well. Both CEO Antonio Neri and head of networks, Rami Rahim, spoke about the importance of networking as a foundation for modern enterprise architecture, particularly for AI workloads.

Rahim offered a cautionary tale about the Millennium Tower in San Francisco, a residential high-rise that began sinking into the ground and listing to one side shortly after its completion due to problems with its foundations. His argument: if businesses don't lay solid foundations based on networking technology, then they're setting themselves up for failure. Their agentic AI ambitions, in particular, will become weak and start to wobble.

Such was the emphasis on networking all week, you would be forgiven for thinking that the company had moved away from its traditional business areas. In fact my first question to Antonio Neri when I spoke to him was: "Is HPE a networking company now, then?"

"Yes, it is," came the answer. He expounded on how this plays into HPE's expectation that it will grow into an $11 billion company, adding that "in my mind, together with our competitor, Cisco, [HPE has] the most complete networking portfolio".

"Our competitor Cisco" – what a phrase, and one that would have seemed slightly crazy just a few years ago.

It's not that HPE has had no networking business before this – Aruba Networks has been a core part of what HPE has to offer for years – but I don't think anyone would have openly declared Cisco a competitor in 2015.

Undeniably the Juniper acquisition does strengthen HPE's position in the data center – it can now offer the whole package, from scale up to scale out. Neri was keen to point out the AMD Helios, which should be coming this autumn, uses HPE Juniper Networks for scale up. Possibly more to come on that in July when we get to hear from AMD's CEO Lisa Su at AMD Advancing AI summit, though we shall see.

The question that I've been asking myself throughout the conference and I've voiced already on the 19 June edition of the ITPro Podcast and reaction video, is this: is the focus on networking a perpetual state of affairs, or is HPE just burnishing its latest acquisition?

My expectation is that for the next 6 months minimum Juniper's star will very much be in the ascendence at HPE. After that? Well, maybe the foundations will have settled and networking will have equal importance to the rest of what the company does. HPE Discover Barcelona, the company's European conference that takes place 2-3 December this year, may offer some more insight into the direction of travel.

Until then, you can catch up on all my coverage from Las Vegas here and if you want to know more about the tower and its ongoing issues – including attempts to remediate that made it worse – I recommend this blog, and also the Wikipedia page. It makes for quite a read.

If an organization is surprised to see IT costs explode when it starts running agents at scale, the problem might not be in the IT budget – it might be a category error. At least in the opinion of HPE CEO Antonio Neri.

“I don’t think of [AI agents] as an IT cost,” Neri tells ITPro at HPE Discover 2026. “I think about the cost of the workforce because, to me, an agent is no different than any other employee I have to hire.”

“So if I hire [a person], there is a cost to onboard him and make sure that he's productive by giving him the tools and everything else,” he says. “When I train an agent, it's going to cost me a number of tokens to train it to drive the best productivity. If I'm going to spend a million dollars to train an agent, it has to be way more productive than [the person]. Otherwise, why am I doing that?”

This turns on its head much of the conversation around agents, tokenmaxxing, and costs as agentic AI starts to eclipse generative in organizational use.

AI sticker shock

ITPro wrote in May about “tokenomics” and so-called sticker-shock when the AI bill finally came due for many businesses. Famously, Uber used up its entire annual AI budget in four months and has now capped the amount each employee can spend on agents per month.

One answer that’s been given to the problem is a shift to on-premises IT rather than ‘cloud-based’ AI like Anthropic’s Claude Code or OpenAI’s Codex, and moving agent costs from OpEx back to CapEx. For an enterprise hardware company like HPE, surely this should be an opportunity to increase hardware sales?

“Sure,” says Neri. “But we think about it as a total cost of [the] workforce, not as an IT expense.”

And is this something that he’s educating his customers about, or do they see it another way?

“Yes, I am,” he says. “I’m spending more than 50% of my time with our customers, and when I bring that concept to life for them, they say ‘ok, ok!’”

“I’d rather give a little bit more budget to IT to reduce my entire cost of the workforce, which is – depending on what you do – probably one of the biggest cost levers,” he continues. “I have only two large cost drivers: people and materials. The materials, in networking, we have a lot of that under control because we have the IP, but a lot of servers, obviously, CPUs, and GPUs I don’t control.

“[The people] I have 65,000 employees. It’s not about reducing employees, which in some cases will happen, but it’s about making those 65,000 employees way more productive with the aid of AI by freeing up capacity for the humans to do more value-added [work] while the agentic models do more work at much bigger scale. I think about that, the total cost of [the] workforce”.

Human / AI interaction

This is an idea that’s been seeded here and there throughout the conference. In Rami Rahim’s keynote, which took place just a couple of hours after Neri’s, one of the demonstrations floated the idea of taking humans out of the loop once a process has been approved.

Katrina Pickett, senior product manager, Juniper Networks, showed how HPE Marvis might help detect and remediate a problem with a sluggish network. The conversation between person and machine was in natural language, but the fixing was agentic. At the point that the fix was applied, Marvis asked if Pickett wanted to have an agent apply the same fix automatically in the future, without human intervention, which would present a time saving.

Human-in-the-loop (HTL), however, has been a key mantra of not just AI but automation more generally. Is it time to take another look at this attitude?

“Well, once you’re confident about [what’s happening], then you can remove the human from the loop and just put on the loop,” says Neri. “I think there needs to be some sort of oversight, but not doing the actual work.”

“In the case of networking, a lot of things were done manually through [the] command line interface, by logging in the console, and putting hands on a keyboard,” Neri continues.

“The scale is now so big that humans cannot possibly manage that scale. Therefore, there is no other choice than to automate things and use artificial intelligence to manage massive amounts of data.”

Naturally, there are caveats. The story of the company that had its entire codebase and all backups erased by a rogue agent is quickly passing from fact into legend; a cautionary tale for any business thinking of letting agents roam free in their network.

“I always say, you don’t get what you expect, only what you inspect [so] I think there needs to be some sort of oversight,” Neri says. “Human beings make mistakes, agents may make the same mistake, right? But as you’re doing it at a bigger scale, it may have a huge impact.”

Mistakes happening are “inevitable”, says Neri. Ultimately, though, there has to be an element of trust that the technology will work as intended. Guardrails can help with this, and, from Neri’s perspective, what HPE is doing with GreenLake Intelligence and Marvis Actions means customers can deploy agents with greater confidence to whatever extent they wish.

HPE still makes everything it always did – it has storage appliances, and it has servers a-plenty, and they’re popular too. According to the company’s most recent earnings call, traditional server orders are booming, growing triple digits year-on-year, while Alletra Storage MP orders also increased triple digits.

The conversation, however, will always revert to networking when it comes to visions of and for the future.

“There was a question in a previous session [that was posed] to Rami (Rahim), which said how far are we to really being fully autonomous, and we believe it’s within the next 12 months,” says Neri.

“It’ s not going to be everywhere across the network, but there are many aspects of the network just running itself.”

“Our opportunities are in the three domains: networking, cloud, and AI. Networking is a huge opportunity; we see it is. And it’s driven both at the edge to the campus and branch, the on ramp into the AI. You need that connectivity.”

Cisco aims to consolidate and simplify infrastructure management to compensate for rapid AI advances and surging customer demands, according to the company’s UK&I CTO.

Speaking to ITPro following the company’s annual Cisco Live event, Rob Lay said a slew of announcements made at the conference last week intend to shore up infrastructure optimization capabilities for enterprises.

Chief among these announcements was the launch of the Cisco Cloud Control platform, which aims to consolidate disparate infrastructure management platforms through a single interface.

The move here, according to Lay, comes in direct response to the growing sprawl of disparate, overlapping solutions enterprises contend with, be that from Cisco itself, or third parties.

“Cloud Control is all about bringing together all of those different management platforms that we've got, because we've got a really broad portfolio, bringing together the likes of Catalyst Center and Meraki Dashboard and Nexus One, and the Collab Command Center, and Cisco Security Cloud Control,” he told ITPro.

“All of those pieces, you’re bringing all that together so you’ve got that single point of management.”

Cisco describes the platform as a “new way” to run critical infrastructure that ties together five key elements, including:

• Cross-domain telemetry
• Purpose-built models
• Trusted agents
• Cloud Control Studio
• Cisco AI Canvas

The launch of Cisco Cloud Control is the culmination of several years’ worth of evolution for the networking giant, according to Lay. Indeed, Cisco has been moving consistently toward a “platform approach” when it comes to network and infrastructure management.

“We've talked for a couple of years now about this idea of platform, and I think that it becomes very, very difficult to really have a true platform until you've got that consolidated management capability,” he added.

While in previous years this focus was spurred on by evolving cloud demands, particularly the shift toward hybrid and multi-cloud setups, the advent of generative, and more recently agentic, AI required somewhat of a rethink.

The performance demands of agentic AI are huge, Lay told ITPro. This means performance, stability, and consistency of cloud and networking capabilities are now paramount.

Running parallel to the performance requirements, enterprises are also dealing with an increasingly sprawling array of agents working behind the scenes, which again requires an overhaul of visibility and the ability to track, monitor, and orchestrate these autonomous bots.

“I think it's also driven by the landscape that we're seeing from an AI perspective as well. We can't get through a conversation these days without mentioning AI, but when you look at the challenges that our customers are facing – the need to be able to adapt, manage, respond to stuff at much, much faster speeds is driving a necessity,” he explained.

“We can't have customers bouncing from console to console to console anymore. So it's about bringing together all those different capabilities [and] unifying that experience for our customers.”

Cisco’s ‘AgenticOps’ push

Agentic AI certainly is a key factor in why Cisco customers are demanding more intuitive, unified management capabilities in 2026, Lay told ITPro.

Indeed, the Cloud Control Panel represents the underlying foundation for the firm’s ‘AgenticOps’ operating model. The company envisages a future where humans and agents are working together “side by side”, and the features rolled out by the company aim to simplify adoption.

The firm’s Agent Builder tool, for example, enables users to build custom agents for Cloud Control “tailored to their own policies and workflows”. These can be connected to more than 50 third-party platforms, either through native connectors, or the Model Context Protocol (MCP).

Elsewhere, the App Builder also allows enterprises to build and publish applications and workflows for Cloud Control using natural language prompts. This, the company noted, is based on OpenAI's Codex agent.

Tackling AI’s “network problem”

While agentic AI features and customization capabilities are a key focus for Cisco, Lay noted that Cisco’s bread and butter, networking, is the area in which it’s still emphasizing the importance of relying on high-performance capabilities.

Customers have worked their way through the early days of the generative AI boom, which was dominated by chatbots, and with agents, they’re now focusing heavily on delivering clear-cut use cases that “add value to their organizations”.

Throughout this process, he noted, it’s become clear that “AI rapidly becomes a network problem”.

“It's really, really difficult to enable an AI platform to work properly unless you've got that very highly capable network that underpins it all,” he explained.

“It is kind of true to say that Cisco now is delivering that critical infrastructure for the AI era, because regardless of whether we’re talking about the data center network and the need to connect to GPUs, whether we’re talking about the fact that we’ve got customers, service providers, and hyperscalers now building data centers further and further apart - how do we work to that?”

With the previous generation of generative AI focused largely around chatbots, network traffic typically “spiked” in short bursts. With agentic AI always on and running away in the background, that “profile just goes up and stays high in terms of network demand”.

“That puts a very different challenge and requirement on the network to deliver that real consistent high efficiency, high quality network connectivity,” he said.

Lay said these fundamental “network challenge” issues have prompted Cisco to drive networking innovation in recent years to enable it to meet surging customer demand.

Part of the challenge here, he noted, has been emphasizing that customers simply cannot continue “sweating assets” if they’re to have success with agentic AI (and remain secure).

In a recent discussion with a customer, Lay revealed an IT leader boasted about having a Catalyst Switch that had been “up for 12 years”.

“That’s great, right? That shows our kit is reliable and stable, but in today’s world that’s quite a risk for our customers, not just from a security perspective, but fundamentally, it’s not going to support any of the sort of network traffic types that you need for today,” he told ITPro.

Customer mindsets are shifting on this front, Lay added, and this is an area Cisco is keen to capitalize on and keep pace with.

“When they’re looking at refreshing their technology, they’re not just looking for the current in-support version of what they had previously, they’re looking at what do we need our network to do in two years’ time, in three years’ time,” he commented.

“That’s what we need to be planning for and investing for today.”

FOLLOW US ON SOCIAL MEDIA

Like millions of office colleagues, Randy Bresnik, Christina Koch, and Kjell Lindgren swapped a bunch of pictures of their pets in June 2024. Unlike most of us, they worked on the International Space Station at the time, 250 miles overhead. The pictures were sent as part of the Laser Communications Relay Demonstration (LCRD), which achieved speeds of around 1.2 GBits/sec.

To many, the pet pics might not be nearly as interesting as the speed. While most internet users get by with 50-100Mbits/sec, over ten times that much sounds like a technical feat that could be used to upgrade business networks.

Part of NASA's Space Communications and Navigation (SCaN) program, LCRD used the same technology you're using to read this story – lasers. Just like on Earth, the LRDC fires a beam of light blinking on and off to represent the ones and zeroes of binary data, the only difference being the absence of fiber optic cabling.

Just a few years later, the pioneering crew of Artemis II have used a laser communications terminal of their own – the Orion Artemis II Optical Communications System – to send 4k footage back to NASA at 260 Mbits/sec, despite being the furthest distance humans have ever been from the Earth.

Better (and faster) cat pictures aside, how do we deploy and maintain a system like that here on Earth? And what might laser internet do for businesses?

Space-based business networks

The commercial possibilities are potentially far reaching. Using systems like Starlink or Amazon Leo, networks of satellites covering every point on Earth's surface, network operators can provide internet anywhere. In the near future, it might be just as easy for businesses to get superfast broadband in rural areas or on planes in flight as it is in urban areas.

On paper, this could one day make subsea internet cables obsolete altogether – large cloud providers like Google or Amazon might network data centers together around the globe from spaceborne transmitters.

We also have an increasing need for low latency internet service. On the stock market, for instance, where milliseconds count, a laser from space might give traders an edge over a light pulse going through a mountain range or across an ocean.

Laser internet also has the means to offer isolated networks that are much more secure and resistant to hijacking or jamming. Militaries or the government could deploy dedicated laser internet systems to coordinate materiel movements or next-gen intelligence communications, as well as safeguard critical national infrastructure

Terrestrial energy and data networks are also very vulnerable to disasters like earthquakes and hurricanes, and when coordinating relief or emergency services is critical, a robust network delivered from orbit can keep the lines of communication open no matter what state the infrastructure is in on the ground.

Stumbling blocks

We already have satellite internet, but it uses radio signals (RF), and concentrated light like a laser needs direct line of sight with the receiver, which is a different proposition (a series of way stations covering Earth to ensure each one is visible by the next is beyond our engineering and economic means).

The LRDC experiment itself proved how tricky line of sight can be, as David Israel, Near Space Network chief architect and LCRD principal investigator at NASA's Goddard Space Flight Center, explains. "Sometimes the solar panels [on the ISS] can themselves be blocking the signal," he says. "If you just have one telescope using [a] very specific line of sight it's very difficult. We don't have the equivalent of antennas that broadcast in all directions."

Even assuming you have a clear line of sight, a laser beam has a lot to get through here on Earth. You can increase the power of an RF signal and circumvent most of the atmospheric effects – we've been doing it with TV and radio for over a century. Not so for lasers. Cloud, wildfire smoke, general pollution and fog can all scatter and obstruct a light beam, and suddenly your internet will be more like dialup.

That's just down here where we live. As Jason Mitchell, principal technologist for Communications and Navigation at NASA, adds, the atmosphere above our heads is a constantly moving soup. "It's surprising how active the upper atmosphere can be. It can result in dramatic changes in the refractive index of a light beam."

But the most surprising stumbling block is actually scheduling. The LRDC platform was in a geosynchronous orbit – the craft stayed overhead the transmitting and receiving stations in Hawaii and California as the earth rotated.

Other SCaN programs have only been possible while the relevant orbiters are passing overhead, and unless you're Elon Musk with over 9,000 Starlink satellites in orbit, achieving global coverage is a big and expensive problem.

But NASA has put a lot of thought into getting around such constraints. Israel talks about a smart network that sends data through nodes experiencing the best conditions, avoiding clouds and transmitting to another orbiter or ground station instead. All this would be happening at light speed. "Your data bounces around the world until it finds a transmitter that can see the receiver clearly," is how he puts it.

Then there are the methods Israel and Mitchell talk about that maintain or 'clean up' the signal. "You still have issues where you need to get the beam re-centered and refocused because it will be sort of gridded and the phase will be different across the wavefront," Mitchell says. "We have coding and modulation and other methods that take care of a lot [of data fidelity]. If you looked at the variation of the signal it'd look like it was moving around a lot but it can be cleaned up without us losing data pretty remarkably

The future of space-based internet

Legends about NASA's commercial successes with Tang (apocryphal), memory foam and scratch resistant sunglass lenses abound, but the LCRD was created for the sake of scientific discovery.

A precursor mission, Deep Space Optical Communications (DSOC), sent data to the receiver at 257 Mbits/sec for more than 4 hours (while it passed by in the sky), receiving 4.7Tbits of information. Contrast that with Magellan mission to Venus in the 1990s, which returned around 1.2Tbits over its five year mission.

That means DSOC sent nearly four times as much data almost 8,000 times faster – in a single pass. "What new discoveries could be made with such communications capability?" Mitchell asks.

Of course, industry will only be interested if it's worth the investment, and nobody's going to throw 5G towers or phone exchanges away any time soon. So how different is the machinery you need to send pet pictures to the ISS? The transmitter is a laser generator affixed to a telescope at an observatory – not fundamentally different to the LED and modulator sending the data of this website down fiber optic cabling right now.

Even more interesting is when Israel describes the device on the ISS that receives the information, something that enables "bits going in and out" and "a fibre optic connection through amplifier". In other words, a modem. Even more crucially, he says at one point "there was no change to the commercial modem". It sounds for all the world like SCaN is an everyday fibre optic internet connection without the fibre optics.

Israel doesn't think we'll all have dishes on our office roofs any time soon, a passing satellite firing a concentrated laser full of internet traffic at us. However, he can see a place for such tech. "Where it fits is in satellite systems, telecom hub buildings or gateways with the big antenna all pointed at at different satellites," he says.

Laser internet isn't common yet – but it's also not science fiction. If you're waiting for it to make your video conferencing experience better that might take awhile, but for now, it has the capability to supercharge our understanding of the universe around us.

In the last five years enterprise connectivity has gone from being a baseline capability to a strategic imperative, and this shift in focus is being driven in part by AI-related demands.

That’s according to Cisco’s inaugural State of Wireless Report. It found business wireless connectivity now represents a “strategic growth engine” for enterprises, helping to drive innovation and deliver downstream success in technology adoption.

Speaking to ITPro ahead of the report’s publication, Cisco Wireless CTO Matt MacPherson said businesses have ramped up investment in connectivity in response to a number of overlapping factors.

“If you go back a few years, it was all about connectivity,” he said. “If you get on the network, that’s what drives productivity, and the more you could get connected to the internet, the more that you have these resources available to you, the gains would be achieved.”

“Now connectivity is sort of assumed, but now it needs to meet the requirements of these next generation applications and services.”

AI workloads, an influx of IoT tools, high-bandwidth applications, and even hot desking have all precipitated huge funding increases and a sharpened focus on wireless connectivity, the report found.

Indeed, 80% of respondents told Cisco they’ve increased investment over the last five years, and 29% have increased budgets by 50% or more across that period.

Looking ahead, they plan to continue in this vein, with a growing number of respondents detailing plans to upgrade to the 6GHz spectrum, and nearly three-in-five firms planning to deploy Wi-Fi 6E or Wi-Fi 7 in the next year to modernize connectivity.

Downstream benefits of network modernization

With AI now dominating the C-suite focus, the impetus to improve connectivity capabilities has skyrocketed, according to Cisco, and for good reason.

The report identified a direct link between increased wireless investment and positive business outcomes, including a return on investment (ROI) in areas such as AI.

More than three-quarters (78%) reported operational efficiency gains from wireless investment, for example, while 75% saw employee productivity improvements. The benefits are also being felt in customer engagement, Cisco found, while 68% reported positive revenue impact.

MacPherson told ITPro that these findings show the penny has dropped for many enterprises with regard to connectivity.

Simply put, modernization efforts in this domain have a “multiplier effect” across the business - especially in AI, and it comes at a critical juncture in global roll-outs.

ROI – or lack thereof – has been a growing concern with generative AI. While recent studies show signs of improvement on this front, many have faced issues with outdated infrastructure.

The demand placed on networks by skyrocketing AI data flows also means enterprises can’t afford to shirk investment on this front. Low latency and capacity are critical in keeping the lights on in this regard.

“I think what’s really clear with some of the data that we pulled in here is the wireless infrastructure underneath is often an afterthought, but absolutely critical for achieving the ROI that people are expecting,” MacPherson noted.

“It’s easy to start loading up the network with some of these new workflows, and then suddenly realize that the network becomes the bottleneck to getting the execution that you’d like to see.”

The “wireless AI paradox”

As in many other areas, when it comes to wireless connectivity AI can be a double edged sword, Cisco found, which the company described as an “wireless AI paradox”.

While AI is helping drive modernization efforts in wireless connectivity and deliver ROI, it also functions as the “principal catalyst of operational complexity and risk”.

“AI has huge promise [and] we’re already seeing big gains – and I think we’re going to continue down that road,” MacPherson told ITPro, “but at the same time, AI can add additional complexity.”

MacPherson used cybersecurity risks as an example of how this paradox is unfolding for businesses. The technology is now being used to automate network management to great effect, the study found.

The vast majority (98%) of those using AI in network management report significant gains, saving an average of three hours and 20 minutes per person, per day.

At the same time, however, threat actors are actively using the technology to automate attacks. Organizations are now engaged in a war of attrition, with AI acting as the primary defensive and offensive weapon.

“Not only does AI speed up the capability to manage a network, immediately see where the issues are in the network, immediately see breaches in the network, and whether those breaches have moved horizontally to cause a more broad problem across your infrastructure,” he explained.

“At the same time you’ve got nefarious forces using AI that can just sit there and keep hammering at your network, whether it's a denial of service type attack, et cetera,” MacPherson added.

Skills shortages are exacerbating the paradox

The third and final aspect of this paradox lies in skills, according to Cisco. Firms are facing significant personnel shortages, which is amplifying complexity and leaving them open to threats.

Nearly nine-in-ten (89%) wireless leaders told Cisco they struggle to hire professionals in networking and connectivity, which in turn is driving 70% higher security incident costs.

Worse still, it’s trapping teams in a vicious cycle of “reactive operations”. Addressing this area will be a key priority for enterprises moving forward, MacPherson told ITPro.

Teams working in this domain need “advanced AI expertise” to navigate the challenges faced in both network modernization, but also in integration of the technology.

MacPherson noted that these talent shortages are akin to previous network modernization periods, particularly the 5G roll-out. Enterprises need time to adapt to changes, and talent acquisition efforts will ramp up accordingly.

“We’ve seen these types of things before when we were doing a lot of work converging, for example, 5G and Wi-Fi networks indoor and outdoors, it was really hard to find someone that was really good at 5G and Wi-Fi and could unify those architectures,” he said.

FOLLOW US ON SOCIAL MEDIA

Commercial 6G networks will arrive “towards the end of this decade”, Ericsson’s head of 6G operations said on the opening day of MWC 2026. And that will see us move from 5G networks that can carry intelligence, to 6G networks that are intelligence.

Speaking on stage at Qualcomm’s EUnion and The Rise of Personal AI event, Marie Hogan explained that the current increase in AI models, data centers, distributed autonomous agents in cars, wearables, and robots has placed a “dramatic impact” on mobile networks.

In particular, the rise of AI agents and intelligent AI-native devices has placed a lot of demand on mobile infrastructure.

“It is not enough just to add more compute; we need more advanced networks,” Hogan explained during her speech. “We need to add more uplink – a lot more uplink, in fact”.

It’s no secret that we are living in a data and information era, and with 5G advanced technology we are starting to see networks that are AI-powered.

For example, last month, Ericsson launched AI-capable radios, built with its own silicon and neural network accelerators as part of a wider AI-first approach.

These remote radios are engineered specifically for higher downlink efficiency and uplink performance.

“We continue to build hardware that’s planning for and enabling AI, but we also are starting to add intelligence to selected functions in the networks,” Hogan said.

“And this is a stepping stone towards the fully capable AI networks that we will see when it comes to 6G. Towards the end of this decade, we will see commercial 6G networks.”

The difference with AI-native 6G

There is a fundamental difference in how networks will be built; with 6G or AI-native 6G, as Hogan called it, AI will be built as a fundamental principle of the network for the first time.

“We will see AI embedded in every layer of the network, from the physical layer all the way up to autonomous control, all the way up to service orchestration, for example,” Hogan said.

“We will see a network architecture that will bring together compute, connectivity, and sensing. What we’re aiming to do is to build architecture is both intelligent and enabling the fabric of the world around us.”

Essentially, the idea is to have an intelligent network with 6G that can interact and understand the world around us, so that it can optimize accordingly.

“As we move from 5G, you can see it as a network with 5G that can carry intelligence to a network in 6G that actually is intelligent,” Hogan said.

FOLLOW US ON SOCIAL MEDIA

Hewlett Packard Enterprise (HPE) has unveiled a series of new solutions aimed at streamlining AI infrastructure networking at the edge.

Announced ahead of MWC 2026, the new networking products mark the latest milestone in the company’s efforts to integrate Juniper Networks products across its infrastructure portfolio.

“This integration builds on the combined strengths in large-scale service provider infrastructure of both HPE and Juniper Networks, along with deep expertise across networking, compute, security, and cloud,” the company said in a statement.

Part of the update will include three new additions to the Juniper PTX router range, which HPE said will deliver a 49% improvement in power efficiency compared to previous generations.

Rami Rahim, executive vice president, president, and general manager for networking at HPE, said the new PTX routers will enable customers to meet surging data traffic demands as a result of AI workloads.

“AI infrastructure is a critical growth driver for service providers. More than ever, HPE is committed to helping these customers lead in the AI era by building intelligent, next-generation networks that can support complex operations, rising data traffic, and the transformative capabilities of AI,” he said in a statement.

Here’s what customers can expect with the new updates

Under the hood of Juniper PTX

The first of the new product launches includes the Juniper PTX12000, a line of modular routers which HPE said will allow network operators to “scale AI and cloud traffic without repeated infrastructure redesigns”.

The new router features an “ultra-dense” 800G port density - on 1.6T-ready platforms, according to HPE. The firm noted this will allow users to sustain “predictable” low-latency performance that can scale based on growing demands.

“The platform scales to 345.6T with the 8-slot PTX12008 and 518.4T with the 12-slot PTX12012, enabling seamless capacity expansion while preserving architectural consistency and reducing latency, power consumption, and costs,” the company said in an announcement.

Meanwhile, new Juniper PTX10002 fixed form routers also aim to match growing traffic demands. This line offers high-density routing designed specifically for AI network fabrics, HPE noted.

The PTX10002 offers 28.8T or 14.4T of throughput in addition to flexible multi-rate port options beginning at 100G through to 800G.

“The expanded portfolio enables efficient scaling of AI clusters and WAN connectivity, while giving customers the flexibility to optimize space, power, and port density as AI and cloud traffic grows,” HPE noted.

HPE teases an “agentic-ready” Routing Director

Juniper Routing Director, a tool used to automate network traffic, is also set for a big “agentic” update as of HPE’s upgrade plans.

This will allow enterprises to connect their own AI assistants, or “copilots”, to the service in a bid to automate WAN routing issues, for example.

HPE said the update will give customers the “flexibility to use agentic AI to automate workflows, optimize WAN routing performance, and simplify AI-native, post-deployment operations”.

Elsewhere, HPE announced plans to integrate the Juniper Cloud Native Router (JCNR), the networking giant’s container orchestration software.

This is now available on 1U HPE ProLiant DL110 servers, as well as the new 2U HPE ProLiant EL140 Gen12 server range.

“This enables service provider operators [to] eliminate routing hardware at cell site locations, reducing hardware purchase expenses and ongoing energy costs,” the company said.

FOLLOW US ON SOCIAL MEDIA

Cloudflare experienced a major outage on Friday that led to issues for a number of websites using its services.

According to Cloudflare, the issue affected customers using its Bring Your Own IP (BYOIP) service, who saw their routes to the internet withdrawn via Border Gateway Protocol (BGP).

"For some BYOIP customers, this resulted in their services and applications being unreachable from the Internet, causing timeouts and failures to connect across their Cloudflare deployments that used BYOIP," the company said in a blog post.

The impacted BYOIP customers first experienced a behavior called “BGP Path Hunting”, whereby end user connections traverse networks trying to find a route to the destination IP - and carry on until the connection that was opened times out and fails.

Meanwhile, visitors to the website for Cloudflare’s recursive DNS resolver (one.one.one.one) were met with HTTP 403 errors and an “Edge IP Restricted” error message.

DNS resolution over the 1.1.1.1 Public Resolver, including DNS over HTTPS, was not affected.

In total, the incident lasted six hours and seven minutes.

Cloudflare outage caused by configuration changes

Cloudflare confirmed the issue was caused by a change the company had made to the way its network manages IP addresses onboarded through the BYOIP pipeline. This change caused it to unintentionally withdraw customer prefixes.

"The specific piece of configuration that broke was a modification attempting to automate the customer action of removing prefixes from Cloudflare’s BYOIP service, a regular customer request that is done manually today. Removing this manual process was part of our Code Orange: Fail Small work to push all changes toward safe, automated, health-mediated deployment," the firm explained.

"Since the list of related objects of BYOIP prefixes can be large, this was implemented as part of a regularly running sub-task that checks for BYOIP prefixes that should be removed, and then removes them. Unfortunately, this regular cleanup sub-task queried the API with a bug."

While Cloudflare moved fast to reverse the change, around 1,100 prefixes were withdrawn first - around a quarter of the total number of BYOIP prefixes. Some customers were able to restore their own service by using the Cloudflare dashboard to re-advertise their IP addresses.

The outage affected a number of websites, including Uber, Workday, Minecraft, Wikipedia, and Microsoft Outlook.

Betting site Bet365 posted on X, "Hi, we’re aware of an issue with our Website/App, and our Technical Team are working to resolve it as soon as possible. We apologise for the inconvenience."

Cloudflare looks to prevent future issues

Cloudflare said it is already working on improving the Addressing API's configuration change support through staged test mediation and better correctness checks.

"Following our Code Orange: Fail Small promise to require controlled rollouts of any change into Production, our engineering teams have been reaching deep into all layers of our stack to identify and fix all problematic findings," it said.

"While this outage wasn't itself global, the blast radius and impact were unacceptably large, further reinforcing Code Orange: Fail Small as a priority until we have re-established confidence in all changes to our network being as gradual as possible."

FOLLOW US ON SOCIAL MEDIA

Openreach is urging UK businesses to get a move on in preparing for the public switched telephone network (PSTN) switch-off in less than a year’s time – and targeted price hikes could be the key to forcing action.

More than half a million business lines still haven’t migrated, the connectivity provider warned, leaving them liable to significantly higher prices and eventual complete loss of service.

While many communication providers have already migrated much of their customer base, a number of smaller or specialist providers have been slower to act – and Openreach said businesses remaining with these providers could face risks tied to equipment failure.

Meanwhile, beyond phones, critical hardware such as fire alarms, burglar alarms, and payment terminals will need to be migrated.

Figures revealed by Openreach show that more than 12,000 lift lines and around 500 lines serving CCTV networks, for example, still need to be upgraded.

"There’s no time left to stall," said James Lilley, director of All-IP at Openreach. "We’ve spent the last year ensuring telecare customers can be migrated safely through our 'Prove Telecare' service, removing the final barrier to the switch-off."

Openreach price hikes look to encourage customers

To help push businesses into making the change, Openreach is planning a series of price hikes for legacy Wholesale Line Rental (WLR) products.

The cost will go up by 20% on April 1, by another 40% on July 1, and by another 40% on October 1 – effectively doubling the rental cost of legacy lines compared to 2025 rates.

"The PSTN analogue network is obsolete, becoming harder to maintain and significantly more expensive to run. We are passing those costs on to providers who continue to sell legacy products," said Lilley.

"If your business is still on this copper service, you will start to pay a premium for a service that will be switched off in 12 months. Most major communications providers moved their customers to digital long ago. If your provider hasn't contacted you, you need to ask why."

Making the switch

Openreach said businesses should start by reviewing their connectivity estate to identify any equipment still relying on the PSTN – and contact their service provider if they're not sure. They can also test equipment for free at Openreach’s test labs.

The connectivity provider said they should make the switch as soon as possible, and is offering a range of resources to help businesses through the transition, including support for edge cases and temporary solutions if needed.

The switch from the PSTN to digital services is expected to be complete by January 2027, but is causing concern. While copper phone lines can be powered from an exchange, internet-based digital phone equivalents cannot.

In the event of a power cut, therefore, access to phone services – including the emergency services – could be cut.

While service providers are dishing out battery backup (BBU) devices, free to vulnerable customers, these have a minimum battery life of just an hour.

FOLLOW US ON SOCIAL MEDIA

AT&T is forging closer ties with Amazon Web Services (AWS) and Amazon Leo – the company's low Earth orbit satellite network – to offer fixed broadband services to businesses.

As part of the deal, AT&T revealed it will migrate workloads to AWS infrastructure and will work closely with Amazon Leo to extend networking capabilities for customers across the US

AWS is also set to benefit from the extended partnership, according to Jan Hofmeyr, AWS VP for telecommunications. The move will see AT&T provide high-capacity fiber connectivity for the cloud computing giant's data centers across the country.

"AWS is working with AT&T to advance their goal of expanding connectivity across the United States," Hofmeyr said.

“AT&T gains access to AWS's cloud capabilities and AI tools to accelerate their infrastructure modernization, while AWS benefits from AT&T's expertise in high-capacity fiber networking,” he added.

“By combining AT&T's leadership in advanced fiber networking with AWS's cloud and AI capabilities, we're creating new opportunities to deliver innovative services to customers nationwide.”

Mutually beneficial for AWS and AT&T

In a statement confirming the deal, AT&T revealed it will migrate workloads from several of its current on-premises technologies to AWS on-prem services as well as the firm’s managed hybrid cloud offering, AWS Outposts.

With the help of AWS Professional Services, the telecoms provider will also use Amazon Q Developer and other AWS Agentic Services to accelerate migration of network service enablement.

According to AT&T, the collaboration with Amazon Leo has dual objectives: bolstering data center connectivity as well as networking capabilities for US-based customers.

Under the terms of the partnership, Amazon Leo will offer internet connectivity services to AT&T, enabling it to broaden connectivity capabilities and deliver fixed broadband services to business customers in areas where such services are needed.

Shawn Hakl, senior vice president for product at AT&T Business, said the partnership marks a “pivotal step forward in shaping the future of connectivity in the United States”.

"Fiber is the foundation of that future – it delivers the speed, capacity, and reliability that modern networks demand,” he said.

“By pairing our expanding fiber infrastructure with AWS’s cloud capabilities, and through our collaboration to deliver the infrastructure of the future by connecting data centers, we’re creating a more resilient, scalable, and intelligent connectivity ecosystem.

What is Amazon Leo?

Launched last November and operating as a separate business, Amazon Leo's mission is to deliver fast, reliable internet beyond the reach of existing networks.

The company aims to provide networking services for an array of customers, ranging from individual households and small businesses to large enterprise and government customers.

Amazon Leo is powered by an initial constellation of more than 3,000 satellites, connected to a global network of ground gateway antennas and dedicated fiber.

This includes a series of compact, high-performance antennas –Leo Nano, Leo Pro, and Leo Ultra – that communicate with satellites passing overhead. The entire system is designed, built, and operated in-house at Amazon.

The project is still in its early stages, with only 180 of the planned 3,200 satellites currently in orbit. However, the long-term aim is to connect tens of millions of customers globally.

Amazon said it plans to start offering services to business customers later this year ahead of a full rollout.

FOLLOW US ON SOCIAL MEDIA

European Trade groups have hit out at the EU’s proposed Digital Networks Act (DNA) amid concerns the legislation could favor “outdated telecom monopolies”.

The European Commission (EC) set out its proposals for the legislation this week, which aims to modernize, simplify, and harmonize regional rules on connectivity networks.

The Act introduces mandatory national transition plans to phase out copper networks and transition to advanced networks between 2030 and 2035.

Meanwhile, the hope is that by cutting red tape, it will encourage operators to invest in rolling out advanced fiber and mobile networks, and that this will in turn attract investment in AI and the cloud.

"High-performance resilient digital infrastructure is essential in strengthening Europe’s leadership in innovation, competitiveness and digital sovereignty," sad Henna Virkkunen, executive vice-president for tech sovereignty, security and democracy.

"Advanced and accessible connectivity will allow start-ups to harness the potential of AI, enable doctors to care for patients remotely, quickly and safely. Our goal is a digital environment where new technologies are readily available, affordable, and grounded in fair trustworthy rules that benefit people."

Key rules under the Digital Networks Act

Under the proposals, companies will only have to register in one EU country to provide services across the union. There will be incentives for pan-European satellite communication services through an EU-level, as opposed to national level.

The Act will also make national spectrum authorization more consistent by giving operators longer spectrum licenses and by making licenses renewable by default.

This, lawmakers said, will ensure that all available spectrum is being used by imposing a "use it or share it" rule on operators.

Similarly, this will also see the introduction of a voluntary cooperation mechanism between connectivity providers and other players, such as content application and cloud providers.

Opposition is mounting

The proposals have drawn fire from the Computer & Communications Industry Association (CCIA Europe). The Act's voluntary cooperation and dispute management scheme, it said, are unnecessary, as online service providers and telecom operators already cooperate.

The new mechanism would leave the door open for legislative amendments or rulings by national regulatory authorities that could turn it into a binding IP dispute-resolution system, the trade group warned.

That would effectively resurrect widely-rejected network fees, allowing dominant telecoms operators to extract unjustified revenues from popular online services.

“This is not a ‘voluntary conciliation’ procedure, but one that will create new disputes. It risks harming every part of Europe’s connectivity ecosystem," said CCIA Europe’s senior policy manager, connectivity and competition, Maria Teresa Stecher.

“The DNA could have been a real opportunity to simplify EU rules and boost competitiveness. Instead, the Commission now risks doing the opposite – favoring the interests of outdated telecom monopolies over European consumers, innovative online services, and better connectivity. This is not simplification, it is a step backwards.”

The GSM Association (GSMA), a lobby group for the mobile communications industry, echoed the CCIA's concerns over the conciliation process, airing a few additional concerns of its own.

Chief among these are concerns that this will add a new layer of complexity for providers and that aspects of the legislation stem from the "largely outdated" ePrivacy Directive.

"A key priority for the Commission over the past 12 months has been ‘simplification’," it said.

"But we now face the prospect of even more sector-specific rules, additional administrative bodies and reporting obligations, and the retention of unnecessary, redundant or duplicative elements including consumer protection and rules stemming from a largely outdated ePrivacy Directive. This is not simplification, it is complexification."

The proposal will now be presented to the European Parliament and Council for approval.

FOLLOW US ON SOCIAL MEDIA

Businesses that want to monitor absolutely everything on their network will find Paessler's PRTG Network Monitor an appealing choice. Many competing products employ device-based licensing and use a modular approach, which can leave you with a mounting bill for optional extras. But PRTG takes a much simpler approach as it's licensed only by the number of active sensors.

A sensor allows a single resource to be monitored, which could be anything from a CPU core, a network interface, or service to one port on a network switch or firewall. Sensor licences can be very cost-effective when you want to control precisely what individual resources are to be monitored.

This approach means there are no hidden costs, as a subscription enables access to all sensors. PRTG comes into its own if you want to monitor virtualization hosts, applications, and even NetFlow feeds, as some vendors only offer these features as chargeable add-on modules.

Paessler operates a rolling update process so as new sensors and features become available, they're yours for the taking. PRTG has a sensor for every occasion with the latest v25.4 on review, offering 321. However, more importantly, it introduces a new user interface, which is currently in alpha testing.

Paessler PRTG Network Monitor 25.4 review: Deployment and discovery

We've been running PRTG in the lab for over eight years and recently migrated it to a repurposed HPE 1U rack server running Windows Server 2022. Hardware demands are modest as the eight CPU cores and 16GB of DDR4 in our host system can handle up to 5,000 sensors, which is sufficient for monitoring around 500 devices.

Those new to PRTG can choose to install it on-premises or save on hardware costs with Paessler's cloud-hosted version. Paessler offers a 30-day unrestricted trial, or you can try it out for as long as you like with a free version that enables 100 sensors.

Either way, you'll find deployment a breeze as a network discovery wizard identifies all devices and assigns a base set of sensors to them, along with pre-set threshold triggers for automatic alerting. Paessler offers an extensive range of monitoring choices as you can access the PRTG core server via its web console, use the Windows or Mac desktop apps, and choose from free iOS and Android mobile apps.

The web console home page presents two donut charts providing insights into used sensors and alarms and even a quick glance reveals how quickly they can get used up. Our lab setup includes a firewall, a 52-port switch, one VMware and three Hyper-V virtualisation hosts, six physical Windows Server systems, two NAS appliances, and a network printer, which combined to slurp up over 700 sensors.

The good news is you can review the components you want to monitor, delete the sensors you don't need, and return them to the pool for use elsewhere. This makes PRTG a lot more flexible than products that employ device-based licensing.

Paessler PRTG Network Monitor 25.4 review: Monitoring and alerting

All monitored items are viewed from the Devices page, and the main tree view automatically organises all systems into hierarchical groupings where they inherit settings such as discovery schedules and login credentials from their parent group. This can get very busy for larger networks, but the view can be customised for your environment by creating new groups and moving monitored systems over to them.

It's easy to identify problems as all sensors are assigned colour-coded icons showing if they are up, down, paused, or in a warning state while sensor pop-up windows provide live graphs and alert messages. Selecting a device or sensor takes you to a detailed overview with live graphs and charts showing activity for the past two days, month, and year.

You can add extra sensors to a device by selecting it and choosing the Recommended Sensors option below its metrics table, or add them manually by using the search facility in the main Sensors page. We had no problems adding PRTG's Redfish sensors to our iDRAC and iLO devices, which gathered detailed information on server health, power supplies, and storage.

PRTG uses sensor state and threshold triggers to perform actions such as executing a program and sending out alert notifications using a wide variety of methods, including email, SNMP trap, SMS, or Syslog, and linking up with other services, including Microsoft Teams, ServiceNow, and Slack. You can set values for sensors at any level of the group hierarchy, and options include state or speed changes, threshold breaches, and traffic volume levels.

Paessler PRTG Network Monitor 25.4 review: New user interface

You can enable the new UI from the console's setup page, and it's clear this isn't just an update, but a complete redesign. As it's in early alpha testing, the home page opens with a bunch of cards highlighting the new features and offering quick links to try them out.

All menus have been moved to the left side for easier access and selecting the Monitoring option presents a new window in the main pane that bears many similarities to PRTG's desktop app. It presents the same split screen view so you can browse through monitored items in the left pane, select a device or sensor, and view all its metrics in the pane to the right.

The main view sees substantial changes as it provides a list of all status icons across the top so you can filter the view to show specific types, such as only those that are down. If you've selected a device, a ribbon menu below these has options for displaying an overview, the associated sensors, logs, settings, and triggers.

Selecting a sensor changes the ribbon menu to show an overview, but with options to view channels and a set of linear performance graphs. The web UI has also been updated so it auto-scales its size when loaded on a mobile device, but if you prefer the older UI, you can swap back to it with one mouse click.

Paessler PRTG Network Monitor 25.4 review: Is it worth it?

Paessler offers a range of affordable subscriptions, with the starter PRTG 500 pack costing €137 per month, paid yearly, which should be sufficient to monitor up to 50 devices. You can easily upgrade to larger sensor packs when required, and the cloud-hosted version also looks good value with a Hosted 500 subscription only costing a little more than the on-premises option.

You will need to tame its sensor usage, but Paessler's PRTG is a highly versatile network monitoring solution capable of providing a wealth of information about devices, applications, and services. All sensors are included in your subscription, so you don't need to worry about optional modules, and the new web interface promises to make it even easier to use.

San Jose, California-headquartered Cisco dates back to 1984, when Stanford University couple Sandy Lerner and Leonard Bosack commercialized an advanced gateway server for local area networking (LAN), launching Cisco Systems.

By 1995, most organizations boasted a Cisco router or two. Cisco's still expanding the market under chairman and CEO Chuck Robbins with releases for the Internet of Things (IoT), AI, and cybersecurity implementations.

With some 90,400 employees, Cisco reported revenue of $56.7 billion in 2025, up 5% year over year. Networking sales rose 12%, enjoying "double-digit" growth for its fourth quarter in a row.

Rodney Clark, senior vice president, partnerships, SMB and mid-market, is the channel head. Cisco lists some 39,971 partners globally on its Partner Locator, including names like Logicalis, Aspire, Comstor, and ePlus.

Key offerings

Cisco emphasizes AI's potential to transform technology markets and customer organizations; multiple 2025 releases target cross-architecture sales with three foci:

• AI-ready datacenters, supporting complexity, cost, and scalability.
• Future-proofed workplaces, including securing and connecting hybrid, distributed networks.
• Digital resilience for compliance and security, including using AI to solve problems faster.

Supporting technologies include Cisco Secure Firewall 200 and 6100 series hybrid mesh firewalling, as well as universal zero-trust network access (ZTNA) and AI Canvas generative user interface (UI) and agentic AIops. A unified platform for Meraki and Catalyst is also expected to drive gains.

Recent news

• June saw Cisco update network security, including for hybrid mesh and network access, to help secure enterprise AI agents.
• Cisco in February said networking products, including integrated platforms, can help drive AI-market sales.
• Last November, Cisco sketched the ongoing 360 Partner Program overhaul as meeting AI workload, infrastructure modernization, and cyber-resilience needs

Cisco for partners

Cisco 360 looks to reward partners, especially for lifecycle and managed services capabilities, answering market shifts. "Enterprise networks are being pushed to their limits," Clark notes.

"Organizations rely more heavily on bandwidth-intensive, latency-sensitive applications like real-time collaboration, ERP, and CRM."

Partners contribute "approximately 90% to Cisco's overall bookings" with software and services slightly behind hardware. So input and suggestions from partners are crucial to Cisco 360's design; support so far includes $80m for AI skills enablement, as well as investment into partner capabilities and branding, suggests Shannon Leininger, Cisco's vice president, global partner sales.

"We know that customers need cross-architecture solutions that deliver AI-ready datacenters, future-proofed workplaces, and digital resilience," she says.

Partner programs, tiers, and partner types

• Under Cisco 360 are Integrator, Provider, Developer or Advisor partner types, covering direct value-add resellers (DVAR), distributors, SIs, ISVs, OEMs, and hyperscalers.
• Preferred Partner designation for top performers, with associated privileges.
• Gold, Premier, and Select tiers end on 25 January 2026. New designations will reflect a "value index" of partner portfolio performance per portfolio, such as security or networking.
• Prior partner investments to receive foundational, capabilities, performance, and engagement credits.
• Specializations will also be revised away from the current system of Advanced or Master competencies reviewed annually, toward solutions and, for Preferred Partners, 'next generation' specializations announced at the November 2025 Cisco Partner Summit. The current system includes:
  • Architecture specializations. Covering collaboration architectures (7), networking architectures (3), and Cisco Umbrella for MSSPs (1).
  • Business specializations. Either in Environmental Sustainability or Customer Experience.
  • Solution specializations. In AI readiness (1), applications/observability (2), hybrid work (1), security (6), or infrastructure (8).
  • Cisco Powered Services managed services specializations. In hybrid work (5), security (6), infrastructure (10), and applications/observability (1).

Training and certifications

• Certifications (14) in Cisco's training catalog comprise 154 courses. Certifications can be tracked via the Alpine certification tracking system. Candidates may choose e-learning on demand, instructor-led courses, bootcamps, or a Cisco U learning path.
  • Cisco Certified Network Associate (CCNA).
  • Cisco Certified Network Professional (CCNP). In Collaboration, Data Center, Enterprise, Security, or Service Provider domains.
  • Cisco Certified Field or Support Technician (2).
  • Cisco Customer Success Manager Specialist
  • Cisco Meraki Solutions Specialist
  • Cybersecurity Associate or Professional (2)
  • Devnet Associate or Professional (2)

Additional training opportunities

• Cisco U offers 200 learning paths and 2,750 courses as well as labs, tutorials, and the like.
• Cisco Networking Academy. For certification and training in topics such as digital literacy, English for IT roles, AI, or advanced programming.
• Black Belt Academy partner training hub with hands-on labs, demos, and role-based learning paths for pre-sales, sales, technical, and engineering roles.

Further benefits, rewards, and incentives

• Cisco Partner Incentive for new customer acquisition, upselling, and cross-selling. Rebates are based on total contract value, customer onboarding completion, and annual increases in contract value and potential bonus rewards.
• Deal registration, not for resale (NFR) products, and demo or lab equipment for reaching certain performance or sales targets. Higher rewards for portfolio use cases.
• Cisco Partner profitability estimator. Launched June 2025.
• Marketing Velocity co-marketing program. Includes customizable campaign kits for partners, on-demand generation, and training.
• One Cisco initiative fostering networking, security, observability, and collaboration, including for Splunk.
• Cisco Seller Rewards for partner salespeople.
• Lifestyle Incentives.
• Partner promotions. Specific regional and technology programs.
• Partner Experience Platform (PXP) with partner value indices, heat maps, data downloads, and more.
• Cisco Networking App Marketplace (previously Meraki).
• Cisco Software Smart Licensing and centralized Cisco software management.
• Mentorship, including through Cisco Global Partner Solution Advisers.
• Cisco branding and badges for advanced achievement in specific portfolios.
• Partner finder.
• Online guides, including the Cisco 360 partner program ebook
• Trials and demos. Or log in for Cisco's dCloud demo platform.
• Talent Bridge partner recruitment matching engine.

Sign up and get on board

Look online to see where you fit, or simply choose and speak to a distributor on the partner locator tool. "A lot of our partners get onboarded through distribution," confirms Leininger.

Online, you'll be taken to this link to create a Cisco account first off, or you can do so here. Portal registration is essential for partner self-service, including finding information, such as contacts, offerings, and partner tools: here's a summary.

Problems? Contact web-based support or email askpartnerprogram@cisco.com.

For help with registration, renewal, or partner access specifically, you can also click here.

For Black Belt Academy support, go here, or log in and navigate to this link.

Other Cisco contacts and support phone numbers can be found here.

Cisco partners can be found on Facebook, X and LinkedIn. A partner community group is here, and partner-focused blogs are here. Cisco also publishes on Instagram, YouTube, and this video portal.

Russian-backed hacker groups are exploiting misconfigured edge devices – now preferring that tactic over hunting down traditional vulnerabilities to gain access to company networks.

That's according to an end-of-year report by Amazon Threat Intelligence, which included details of a nearly five-year-long campaign by Russian state-sponsored hackers that AWS said marked a pivot in tactics.

Now, the primary initial access vector has become misconfigured customer network edge devices, according to AWS Security CISO CJ Moses, with a decline in traditional vulnerability exploitation.

"This tactical adaptation enables the same operational outcomes, credential harvesting, and lateral movement into victim organizations’ online services and infrastructure, while reducing the actor’s exposure and resource expenditure," Moses noted in a post on the AWS security blog.

Amazon isn't the first to spot a shift to edge devices. Sophos noted in April that edge devices like firewalls and routers were the main initial attack vector for 30% of incidents last year, while governmental security agencies issued guidance about vulnerabilities in edge devices back in February.

Last year, Fortinet's FortiGate edge devices were exploited by China-backed hackers, and the FBI warned that Ubiquity EdgeRouters were targeted via a wide-ranging hacking campaign.

How the edge campaign unfolded

According to Amazon, hackers targeted enterprise routers, VPN concentrators and remote access gateways, network management appliances, and cloud-based project management systems, as well as collaboration and wiki platforms.

Moses said the evidence suggested that the hackers were using packet capture and traffic analysis to target network edge devices. This involves attackers compromising an edge device hosted on AWS, capturing packets to harvest credentials, and then using those to penetrate an organization's services or infrastructure, gaining access to wider systems.

AWS telemetry suggested the attackers maintained persistent connections, allowing for data to be stolen.

"Targeting the 'low-hanging fruit' of likely misconfigured customer devices with exposed management interfaces achieves the same strategic objectives [as vulnerability exploitation], which is persistent access to critical infrastructure networks and credential harvesting for accessing victim organizations’ online services," Moses added.

Notably, the company revealed those attacks included customer network edge devices hosted on AWS.

"This was not due to a weakness in AWS; these appear to be customer misconfigured devices. Network connection analysis shows actor-controlled IP addresses establishing persistent connections to compromised EC2 instances operating customers’ network appliance software," Moses noted.

That said, AWS did take action. For example, the cloud giant notified affected customers that they were compromised, fixed compromised EC2 instances, and alerted network alliance vendors.

"Through coordinated efforts, since our discovery of this activity, we have disrupted active threat actor operations and reduced the attack surface available to this threat activity subcluster," Moses added.

To protect edge devices against this style of attacks, organizations must secure and monitor their edge devices, Moses advised, keeping watch for unexpected packet capture files or utilities and enforcing strong authentication – in particular those working in the energy sector or critical national infrastructure.

A 'clear evolution in tactics'

This particular string of attacks dates back to 2021 and appears to be associated with Russia's Main Intelligence Directorate, Moses said.

Similarly, the campaign focused primarily on critical national infrastructure in the West, targeting the energy sector and its supply chain in particular.

Moses traced a timeline back to 2021's WatchGuard exploitation, saying that's when Amazon first spotted the hackers targeting misconfigured devices.

That continued in the next few years with misconfigured devices targeted alongside exploits such as the Confluence and Veeam flaws, but by 2025 the hackers' success with edge devices led to that taking centre stage with zero-day exploitation activity declining.

"The campaign demonstrates a clear evolution in tactics," Moses said.

FOLLOW US ON SOCIAL MEDIA

HPE Discover Barcelona 2025 was in full swing this past week, with thousands of attendees descending on the Fira Barcelona to hear the latest news on the networking, servers, storage, supercomputing – and, of course, AI.

It’s a pivotal time for the firm, as it consolidates its hardware partnerships and heralds a recent acquisition, while laying out its strategy to help customers not only meet demand, but expand their networks and adopt new technologies.

What are some of the biggest things HPE announced – and what does the firm have lined up for 2026 and beyond?

In this episode, Rory interviews Jane live on the ground at HPE Discover Barcelona 2025, to unpack all things HPE.

Highlights

"Yes, the integration is rapid, and that's something that both [Rami] Rahim and [Antonio] Neri have been pointing to, that they're not going to take years and years to do this, that, you know, they're starting it now."

"So what's happening? I mean, calling it a 'cross pollination' is perhaps, in my opinion, not quite the right word, because what's actually going on is sort of moving bits from each company across to effectively the other company. So while HPE is talking about that eventually they'll just be unified, and it's going to be one networking business, as things are there is an understanding of the fact that Juniper and Aruba, their backgrounds and specialties are quite different."

"So on the Monday, Nvidia and HPE announced that the first AI factory lab in the European Union – important caveat – is going to be installed in the city of Grenoble in the south of France. It is focused on the sovereign AI element of this. Now, sovereign AI is something that frequently – sovereign cloud as well – frequently comes up at the European conferences, because realistically, it's Europeans and the Middle East who really care about this stuff."

"HPE has launched a Helios project with AMD. So Helios was announced at advancing AI in June, it is a double wide rack scale AI solution. It's being positioned that HPE is the first to do this and it was kind of hinted at that this joint announcement from them, it's Q1 announcement, which is largely what AMD CEO Lisa Su was like hinting at back in June."

Footnotes

• HPE Discover Barcelona 2025: All the news, updates, and announcements live from Barcelona
• HPE promises “cross pollinated" future for Aruba and Juniper
• HPE and Nvidia launch first EU AI factory lab in France
• HPE and AMD reveal first Helios offering for cloud service providers and neoclouds

Subscribe 

• Subscribe to The IT Pro Podcast on Apple Podcasts
• Subscribe to The IT Pro Podcast on Spotify
• Subscribe to the IT Pro newsletter
• Join us on LinkedIn

Several websites are experiencing major errors apparently linked to Cloudflare, with sites such as LinkedIn and Canva seemingly suffering widespread issues.

The outage comes just over two weeks after Cloudflare’s last global outage, which took out a number of popular websites including ChatGPT.

It’s unclear at present whether the issue is a repeat of the technical problem that caused the previous outage, in which a database error in its bot management system led to widespread software failures on its network machines.

ITPro noted websites such as LinkedIn, Canva, and even the outage checker site Downdetector displayed a ‘500 Internal Server Error’ message.

UPDATE: Cloudflare has identified the cause of the issue which it says is now fixed.

"A change made to how Cloudflare's Web Application Firewall parses requests impacted the availability of Cloudflare's network at approximately 8:47 GMT and concluded approximately 9:13 GMT," a Cloudflare spokesperson told ITPro.

"This was not an attack; the change was deployed by our team to help mitigate the industry-wide vulnerability disclosed this week in React Server Components. We will share full details in a blog post today."

The past month has seen a wave of outages hitting online service providers, including one affecting Amazon Web Services (AWS) and another causing widespread disruption to sites hosted via Microsoft Azure.

None of the outages were linked to cyber attacks such as Distributed Denial of Service (DDoS) attacks, but instead were all technical issues.

“Today’s disruption underscores something many of us in cybersecurity and tech have long warned about – as the Internet has grown more complex, a handful of infrastructure providers end up holding unexpectedly large power over its functioning,” said Richard Ford, CTO at Integrity360.

“Cloudflare sits at the heart of that, providing CDN, proxying, routing, DNS and caching so that websites can stay fast, secure and resilient under load.”

Ford said businesses should consider more investment in redundancies and hybrid cloud setups and question reliance on a single provider for their critical infrastructure.

Madison Square Garden Entertainment Corp. (MSG Entertainment) has announced a new multi-year partnership with Cisco, cementing the networking giant as an ‘Official Partner’ of Madison Square Garden.

The partnership aims to future-proof the iconic New York venue’s digital footprint. The collaboration focuses on leveraging a flexible, high-performance infrastructure capable of handling the massive data demands of modern live entertainment, from high-density fan Wi-Fi to back-of-house operational systems.

Powering a connected experience

MSG will leverage a suite of Cisco’s enterprise networking and security hardware. This includes the use of Cisco Catalyst Switches to secure and simplify the network foundation, alongside the Cisco Catalyst Center for centralised management.

According to the technical specifications released, the venue is prioritising automation and visibility. The Catalyst Center provides IT teams with advanced analytics on network performance and usage patterns, allowing for "proactive optimisation and an enhanced end-to-end experience."

Crucially, the infrastructure is being touted as ‘AI-ready’. The Garden utilises Cisco Nexus 9000 series switches, managed by the Cisco Nexus Dashboard, to ensure its data centres can deliver the compute power and speed necessary for next-generation applications.

It’s incredibly important to have robust connectivity for the visitor experience, according to Doug Jossem, executive vice president of global sports and entertainment partnerships at MSG Entertainment.

“Cisco’s expertise in building high-performance network infrastructure enables us to provide seamless connectivity for artists and fans alike,” Jossem said. “We’re proud to expand our relationship with Cisco and welcome them to the MSG Entertainment family as an Official Partner.”

Security and wireless resilience

Security remains a primary concern for high-profile venues. The current deployment includes the Cisco Identity Services Engine (ISE), integrated with the Catalyst Center, to automate secure connectivity across the venue. This allows the technology team to maintain a rigid digital foundation for every event, whether it be a Knicks game or a sold-out concert.

On the consumer side, Cisco Wireless hardware handles the heavy lifting of fan connectivity, a critical requirement as audiences increasingly rely on digital platforms for ticketing, social media, and in-venue apps. The same secure wireless hardware supports back-of-house operations, ensuring digital resilience for the venue’s staff.

The partnership opens the door for further technological integration, according to Rob McQueen, Cisco’s vice president of global sponsorships.

“Madison Square Garden is one of the world’s most iconic venues, and we’re proud to deliver the networking infrastructure that helps power its network technology operation,” he said.

“Through our new partnership, we look forward to exploring opportunities to further integrate Cisco technology across the venue’s digital footprint.”

This deal follows a broader trend of major entertainment venues partnering with enterprise tech firms to address surging data demand and the shift toward smart, connected stadium experiences.

The question of how HPE will integrate Juniper Networks’ technology into its existing networking portfolio has moved a step closer to resolution at HPE Discover Europe 2025.

Speaking to journalists ahead of the event, Rami Rahim – formerly CEO of Juniper Networks, now EVP and GM of HPE Networking – said the business’ goal is to “unify the experience of these two platforms by leveraging the microservices architecture for both and cross pollinating capabilities from one onto another”.

What this means in practice, for now at least, is that both Juniper and Aruba Central will co-exist within HPE. Some of the technologies, however, will be migrated to where the company believes they fit best.

Marvis, a virtual assistant previously described by Juniper as a “key part of the Juniper AI native networking platform”, and Large Experience Models (LEM) will be moving over to Aruba Central. The rationale behind this, Rahim said, is that both these capabilities are AIOps, which is where Aruba Central is strongest.

Meanwhile, client profiling and organizational insights from Aruba Central will move over to Juniper Mist.

Rahim was keen to emphasize that integration “over time” didn’t mean at some point in the distant future, adding that these initial changes will be happening “in the first quarter of next year (2026)”.

“This is just the beginning,” added Rahim. “We're going to continue to do this cross pollination, and as we do this, we're going to bring the best of both platforms to all users. Nobody gets left behind.”

At the hardware level, there’s also a unification strategy underway, with the announcement of a dual platform Wi-Fi 7 access point (AP), which works across both Aruba Networking Central and Aruba Networking Mist.

Describing this as the “first of many” of this type of hardware, Rahim said: “This is a huge breakthrough [with] massive advantages for our customers, starting with flexibility. Dual platform design lets our customers choose their preferred control point, whether it's Mist or Aruba Central, and switch seamlessly between the two platforms as their needs evolve, with no new hardware involved in that decision.”

New Juniper products hit the market

HPE also took the wraps off two new Juniper Networking products at the conference: a new switch – the QFX5250 – and a new multiservice edge router – the MX301.

The QFX5250 switch connects GPUs within data centers. It’s built on Broadcom Tomahawk 6 chips, offers 102.Tbps bandwidth and, according to HPE, is “the world’s highest performance Ultra Ethernet Transport ready switch”.

The MX301 router, meanwhile, delivers speeds of 1.6 Tbps and offers 400G connectivity across diverse environments and use cases, the company said, such as AI inference, mobile backhaul, enterprise routing, and multiservice.

HPE Aruba Networking Central On-Premises 3 also gets a lick of paint with a redesigned user interface. Additionally, the software now also incorporates both “advanced generative [AI]” and traditional AIOps capabilities, as well as proactive remediation, simplified documentation search and what HPE described as “actionable AI alerts”.

Juniper technology has also taken center stage in other announcements from HPE Discover 2025, including in the company’s Helios partnership with AMD and the soon-to-open HPE AI Factory Lab in Grenoble, France.

MORE FROM ITPRO

• HPE and AMD reveal first Helios offering for cloud service providers and neoclouds
• HPE and Nvidia launch first EU AI factory lab in France
• Three things I expect to see at HPE Discover Barcelona 2025 (and two I don’t)

AWS has taken steps to boost resilience weeks after a major outage, rolling out a business continuity feature targeting DNS records.

Last month, an outage at the hyperscaler knocked an array of popular services offline – from Signal and Slack to Zoom and even Amazon itself – taking several hours to fully resolve.

The outage stemmed from DNS resolution issues and took place in what AWS calls US-EAST-1 Region, better known as northern Virginia, where plenty of massive data centers are located.

A month on, AWS has announced a new tool to help mitigate future incidents, helping to get systems back online within an hour after DNS issues, specifically in that often troublesome US-EAST-1 Region.

"Today, we’re announcing Amazon Route 53 Accelerated recovery for managing public DNS records, a new Domain Name Service (DNS) business continuity feature that is designed to provide a 60-minute recovery time objective (RTO) during service disruptions in the US East (N. Virginia) AWS Region," said Micah Walter, senior solutions architect, in a post on the AWS blog.

"This enhancement ensures that customers can continue making DNS changes and provisioning infrastructure even during regional outages, providing greater predictability and resilience for mission-critical applications."

AWS touts new tools

Walter explained that the tool targets DNS changes that customers can make within 60 minutes of a service disruption in that region.

"Customers running applications that require business continuity have told us they need additional DNS resilience capabilities to meet their business continuity requirements and regulatory compliance obligations," he said.

When services are disrupted, the new tool offers several benefits. "The feature maintains access to essential Route 53 API operations, ensuring that DNS management remains available when it’s needed most," Walter added.

"Organizations can continue to make critical DNS changes, provision new infrastructure, and redirect traffic flows without waiting for full service restoration."

However, the company also stressed that wasn't a sign of an issue at AWS, but pinned the need for the tool on demand from customers in sensitive industries.

"While AWS maintains exceptional availability across our global infrastructure, organizations in regulated industries like banking, FinTech, and SaaS want the confidence that they will be able to make DNS changes even during unexpected regional disruptions, allowing them to quickly provision standby cloud resources or redirect traffic when needed," Walter noted.

How the new tools work

The system works with existing setups and customers can continue using their existing Route 53 API endpoint without changes to apps or scripts.

"The implementation is designed for simplicity and reliability,” he said. This means customers won’t need to “learn new APIs or modify existing automation scripts”.

"The same Route 53 endpoints and API calls continue to work, providing a seamless experience during both normal operations and failover scenarios."

In the post, AWS noted that accelerated recovery isn't enabled by default, but switching it on for a new or existing hosted zone simply involves switching it on.

"You can enable this feature through the AWS Management Console, AWS Command Line Interface (AWS CLI), AWS Software Development Kit (AWS SDKs), or infrastructure as code tools like AWS CloudFormation and AWS Cloud Development Kit (AWS CDK)," he said.

Accelerated recovery is available now for Amazon Route 53 public hosted zones and is free.

MORE FROM ITPRO

• The Microsoft Azure outage explained
• Inside a cloud outage
• The Cloudflare outage explained

UK channel partners are at the forefront of AI readiness and are well-positioned to capitalize on fresh revenue opportunities, according to new research from Westcon-Comstor.

In collaboration with Coleman Parks, the distributor quizzed senior IT decision makers at 500 channel partner organizations – including MSPs, MSSPs, VARs, and systems integrators – across the UK, Spain, Australia, Singapore, and the UAE.

The survey found that 43% of UK-based participants now offer advanced, AI-integrated network services compared to the global average of just 26%.

While the remaining majority are not yet being able to design and deliver these solutions, confidence and capability levels are higher among UK partners.

Most respondents (52%) said they are currently in the process of developing their AI capabilities, while just 5% said they were only “exploring” the technology while offering traditional networking services, a figure significantly lower than the global average of 18%.

Customer support

Additionally, almost half (46%) reported that their customers’ current network infrastructure setups are fully prepared and optimized to support AI workloads in contrast to the global average of 33%.

Most are already supporting clients in network transformation areas such as visibility, security, and automation. Active areas include AI-driven network automation (65%), edge computing and distributed data routing (64%), and intent-based networking (52%).

Rene Klein, Westcon-Comstor’s executive vice president for Europe, said the findings suggest that UK partners are ahead of the curve when it comes to integrating AI into their offerings.

“Our research shows that, despite some challenges, UK partners are leading the way internationally in AI readiness,” he explained.

“This confidence puts them in a strong position to seize the significant growth opportunities on offer in AI-led networking as we look to 2026.”

Growth opportunities

According to Westcon-Comstor, end-users continue to report struggles with internal skills gaps and uncertainty around tech decision making, creating opportunities for channel partners to step in and act as strategic advisors in the era of AI.

Within AI networking specifically, partners cited AI-enhanced network security analytics (65%) as the top growth area and revenue driver over the next 12-18 months, followed by AI-powered network observability and diagnostics (64%).

Additional high-growth opportunities include edge-to-cloud data routing and optimization (59%), as well as autonomous and self-healing networks (39%).

“AI is changing the game, and partners have a rare window to step up as true strategic advisors,” commented Patrick Aronson, chief marketing officer and executive vice president, Asia-Pacific, at Westcon-Comstor. “Customers are hungry for intelligent, adaptive networks.

“The partners who lean in now, build capability and get closer to the customer will be the ones leading the next wave of growth.”

MORE FROM CHANNELPRO

• Box and AWS announce new multi-year AI collaboration
• HPE launches first phase of new-look Partner Ready Vantage program
• Pax8 and Microsoft are teaming up to supercharge MSP growth

Web users globally were met with error messages and site crashes yesterday after an outage at Cloudflare brought much of the web to a standstill.

The incident marks the latest in a string of major outages that have sent the web into meltdown this year.

Amazon Web Services (AWS) and Microsoft Azure, for example, both encountered technical faults that caused widespread disruption for consumers and enterprises alike.

While Cloudflare services are now back to normal and websites up and running again, like previous incidents, the outage will have had a serious impact on operations for a wide range of enterprises.

So what happened?

How the Cloudflare outage unfolded

Disruption began at around 11.20am UK time (7.20am EST) and was initially described as an “internal service degradation” which impacted limited services.

It’s around then that web users began reporting serious difficulties accessing a host of popular websites, services, and platforms.

OpenAI’s popular chatbot, ChatGPT, was among the services affected by the incident, along with social media site X and creative design platform Canva.

Online games such as League of Legends and Valorant were also impacted by the incident.

To add insult to injury, Downdetector, a site used to track web outages, was also temporarily taken down by the outage, preventing users from even checking the source of the disruption.

The cause of the outage

First and foremost, it’s important to note that the outage was not caused by a cyber attack of any kind.

In a blog post detailing the incident, Cloudflare CEO Matthew Prince said the company initially suspected that the source of the outage was due to a “hyper-scale” distributed denial of service (DDoS) attack. However, this was quickly dismissed.

Instead, Prince revealed the incident was “triggered by a change to one of our data systems’ permissions”.

While this appears rather vague, the root cause here lay with the company’s Bot Management system, which is used to protect against threats such as DDoS.

This is used by customers to “score” bot activity and control which bots are allowed to access an individual site – or block access.

As Prince noted, this system includes a “machine learning model that we use to generate bot scores for every request traversing our network”.

Thereafter, these scores are held in a “feature file” which refreshes every five minutes to keep pace with the behavior of bots attempting to access any given site.

These refresh cycles are generated by a query running on a ClickHouse database cluster, according to Prince. Changes to that query caused the database to “output multiple data entries”, which in turn meant the file doubled in size.

“The larger-than-expected feature file was then propagated to all the machines that make up our network,” Prince explained. “The software had a limit on the size of the feature file that was below its doubled size. That caused the software to fail.”

Services are back up and running

Recovery efforts by Cloudflare meant that core traffic flows were back to normal by 2.30pm (UK time), with full recovery by 5.06pm.

Some users continued experiencing difficulties in the hours following, but this was largely due to a surge in network traffic as services came back online.

As with previous outages in recent months, some industry stakeholders have used these incidents to highlight the delicate infrastructural balancing act that keeps the modern economy running.

When one system or feature fails, it can have a devastating global impact, according to Brent Ellis, principal analyst at Forrester.

“In this case, the 3 hour, 20 minute outage could have direct and indirect losses of around $250 million to $300 million when you consider the cost of down-time and the downstream effects of services like Shopify or Etsy that host the stores for tens to hundreds of thousands of businesses.”

MORE FROM ITPRO

• UK and Irish businesses "severely underestimating" the cost of IT outages
• Website downtime costs businesses thousands a month
• Manufacturers report millions in losses as downtime wreaks havoc on operations

The AI boom is fueling high demand for compute power and this is leaving enterprises with no choice but to spend billions of dollars on AI infrastructure. Blackstone has estimated that data centers in the US alone will need $1trn in investment by the end of the decade just to keep pace with demand.

“Bigger and better AI requires compute – a lot of it,” Mike Bushong, vice president of data center at Nokia, tells ITPro. “And, for every megawatt of data center capacity deployed today, networking is the second-largest budget item behind the AI systems themselves.”

There’s a good reason for this. Generative AI models rely on thousands of graphics processing units (GPUs) in a single data center to share information across nodes and racks in real-time. If GPUs are brain cells, then a network is the nervous system connecting them all by sending signals.

Unlike traditional workloads, high-bandwidth, low-latency fabrics – these are unified networks of switches – are critical for data to be sent along this nervous system in real-time. Performance and reliability have always been important, points out Bushong, but “they’ve now become the determining factor in the return on AI investments”.

Yoram Novick, CEO of storage-as-a-service provider Zadara, warns that “simply adding more GPUs without ensuring adequate interconnect bandwidth can lead to diminishing returns”. Underinvesting in networking, he adds, can cause “communication bottlenecks can idle expensive compute resources”.

Knowing the networking technologies

At the center of AI networking are a handful of key technologies to know about, namely Ethernet, InfiniBand, NVLink, and Ultra Accelerator Link (UALink).

Ethernet and Infiband are technologies that connect multiple servers. The latter has been the go-to choice, especially for large-scale AI training, because of its ultra-low latency and higher bandwidth. According to research by Dell’Oro Group, the market was dominated by Infiniband in 2023, with an 80% share, but it’s expected to be overtaken by Ethernet in the near future.

Dell’Oro Group expects the majority of switch ports deployed in AI back-end networks will be 800 gigabits per second this year, 1.6 terabits per second (Tbps) in 2027, and 3.2 Tbps in 2029.

Ethernet has been positioning itself to challenge InfiniBand’s dominance. The first version of 800-gigabit Ethernet was released in February 2024, while the Ultra Ethernet Consortium (UEC) has been set up to enable Ethernet to support more complex and intensive AI workloads. The UEC’s founding members include AMD, Broadcom, Cisco, Intel, Meta and Microsoft. The 1.6-terabit Ethernet standard is expected to be finalized next year.

To put it simply, Ethernet and InfiniBand are primarily for scaling out networks. On the other hand, NVLink and UALink are designed for scaling up networks.

Indeed, a number of widely-used, purpose-built networking technologies are based on Ethernet. This includes HPE Slingshot, the firm’s networking fabric technology designed for high-performance computing (HPC) and AI computing capacity, which is used in some of the world’s fastest supercomputers such as El Capitan, Frontier, and Aurora. New HPE supercomputers such as Discovery will also use Slingshot.

To put it simply, Ethernet and InfiniBand are primarily for scaling out networks. On the other hand, NVLink and UALink are designed for scaling up networks.

NVLink is Nvidia’s flagship interconnect technology. Established in 2014, it enables communication between GPUs in a single server, allowing them to share memory and compute. For more demanding AI workloads, Nvidia also offers the NVSwitch technology, which can support multiple NVLink connections.

At the end of last year, several members of the UEC helped to establish the UALink Consortium. The aim is to challenge NVLink’s dominance with a new standard for low-latency, high-speed communication between accelerators and switches.

The inaugural UALInk specification was released in the first half of this year and can support up to 1,024 accelerators connected through UALink switches. For comparison, the NVL72, a server rack that utilizes the latest NVLink technology can connect 72 Blackwell GPUs – up to 576 GPUs can be supported when a maximum of eight NVL72 racks are interconnected.

Enabling the shift from training to inference

Both Bushong and Novick agree that enterprises building and scaling data centers – be that up or out – need to strike a balance between raw compute power and data transfer capabilities. This balance is becoming ever more important as AI workloads transition from training to inference.

Inference is the term used to describe the processing of prompts by AI models, to produce outputs, predictions, or other desired outcomes. In short, it’s a blanket term used for all real-world generative AI use cases, in which AI models actually process data, as opposed to the training of new models.

Much has been written about the intense compute demands of AI training. This is largely true, as training clusters process tens of trillions of tokens including publicly-available data and synthetic data to form the backbone of frontier LLMs.

While inference clusters may not be crunching data, there’s more potential for critical business processes to hinge on them as they power much of the real-world use cases for cloud AI. They need exceptionally low latency in order to handle workloads without impacting user experience, as well as a high degree of reliability even when a huge number of users are sending prompts to be processed.

Right now, training accounts for around 60% of data center expenses, according to Bloomberg Intelligence analysts. But this is expected to shrink to as little as 20% by 2032 as more resources are shifted towards inference. As demand for inference grows, workloads won’t just be run across clusters in a single data center but across multiple data centers.

“Inferencing systems need to be tightly integrated with networking architectures that can handle high volumes of concurrent requests and deliver results with minimal latency,” says Novick.

Bushong adds that the new age of AI infrastructure means that efficiency is defined by networking performance and not compute speed alone. “Enterprises can maximize their investments and gain real differentiation through networks built explicitly for AI,” he concludes.

Netgear has cut the ribbon on its new Drive Partner Success Program, which has been designed to simplify how channel partners do business with the networking solutions vendor.

Integrators, MSPs, and VARs now have access to a host of new streamlined tools, go-to-market resources, and enhanced partner-focused support geared towards driving growth and profitability.

The move marks the latest phase of Netgear’s increased focus on the enterprise, with the channel revamp aiming to strengthen collaboration, simplify engagement, and expand enablement for partners serving SMEs.

Netgear Drive rewards partners for effective service and long-term co-investment through its three tiers – IGNITE, APEX, and APEX MSP – which have been designed around different partner service capabilities and business models, offering benefits based on partner investment as well as completion of training and certifications.

The company said this structure will help its partners “stand out in a competitive market” and better sync with the unique needs of their customers.

Through the program’s partner portal, partners also now have access to a range of additional perks, including co-investment opportunities, training and certifications, marketing materials and collaborative campaigns, as well as partner-focused technical support.

Rollout

Drive will see a global rollout as part of Netgear’s goal to offer a consistent channel experience to its pool of partners around the world.

The new training and certification materials are available now, alongside new partner pricing and program benefits. However, Netgear said existing partners can continue operating under existing terms during the transition period to ensure a smooth onboarding experience.

Netgear Enterprise said the new program aligns with its mission to provide simplicity, reliability, and cost-effectiveness for SMEs and marks the latest phase of its evolution under the leadership of president and general manager, Pramod Badjate.

“The goal of our Drive Partner Success Program is simple: make it easier for our partners to succeed and grow,” Badjate commented.

“Partners are the center of everything we do, and Drive will equip them with the tools, training, and support they need to accelerate business opportunities and revenue growth.”

MORE FROM CHANNELPRO

• Darktrace bolsters expansion plans with double C-suite appointment
• Arctic Wolf celebrates partner growth milestone in the UK and Ireland
• Mimecast names Graham Douglas as chief customer and revenue officer

Amazon Web Services (AWS) is building a new transatlantic subsea cable system linking the US and Ireland known as Fastnet.

Fastnet is set for deployment in 2028 and will link Maryland and County Cork with a line offering more than 320 terabits per second. The cloud computing giant said equates to the entire digitized Library of Congress every three seconds, or 12.5 million HD films at the same time.

The aim of the Fastnet link is to boost capacity but also network resilience, with AWS insisting it will reduce service disruptions for its customers by offering an additional access point in case other cables have issues – a promise that comes weeks after a massive AWS outage caused by a DNS fault.

The cables are part of AWS' growing private network, which began in earnest back in 2016 and now makes up over nine million kilometers of cabling over land and under sea.

AWS noted that unlike the public internet, which requires networking devices to make a routing decision with limited network visibility, its own system uses real-time traffic monitoring for "complete visibility into every link within the global network”.

This, the tech giant said, will help optimize performance and avoid traffic congestion.

AWS eyes scalable networking

AWS said Fastnet makes use of advanced optical switching branching unit technology, a piece of kit that splits traffic via different routes, letting AWS choose the best path by remotely reconfiguring the system.

"This state-of-the-art unit is strategically positioned on the cable route, enabling seamless redirection of data to future landing points as network demands evolve," AWS said.

"This scalable architecture is specifically designed to handle growing AI traffic loads, allowing customers to rapidly expand their data demands while the system adapts to accommodate future growth."

Beyond that, AWS said it has armored the cable with extra layers of steel wires in areas closer to shore to protect against "natural and human activity". That comes a year after two Baltic subsea cables were severed in what appeared to be sabotage.

Notably, the landing point locations in Ireland and Maryland were carefully chosen to offer a route away from traditional cable corridors.

That offers a backup route if other cables have issues, but also lets AWS create a network designed to support growing demands from cloud computing, AI, and edge applications.

Taoiseach Micheál Martin welcomed the announcement, adding that the investment will help “reinforce the resilience of critical digital infrastructure” across the Republic of Ireland.

"By linking County Cork to Maryland in the United States, Ireland will become a true gateway to Europe for submarine telecommunications cables," Martin said.

Wes Moore, Governor of Maryland, added: "Building Maryland’s first-ever subsea fiber cable is an achievement bigger than broadband connectivity — it’s about securing Maryland’s status as a global hub for innovation, job creation, and high-tech investment."

"To grow our economy, we need to prioritize industries of the future," Moore added. "That is exactly what this project helps us deliver, in partnership with leaders with deep local ties and robust commitment to community engagement."

MORE FROM ITPRO

• INSERT CONTENT

Cisco has unveiled a new computing platform designed specifically for AI workloads on the edge.

Announced at the company’s annual partner summit, Cisco Unified Edge will combine computing, networking, and storage capabilities in a single platform to provide “real-time AI inference” and support for agentic AI workloads.

Cisco said the new platform is aimed primarily for organizations operating in data-intensive sectors, such as healthcare, manufacturing, and retail.

The networking giant noted it has “partnered closely” with customers across the aforementioned industries to design the platform.

"Today’s infrastructure can’t meet the demands of powering AI at scale,” said Cisco president and CPO, Jeetu Patel.

“As AI agents and experiences proliferate, they will naturally emerge closer to where customers interact and decisions are made – the branch office, retail store, factory floor, stadium, and more.

“That’s where compute needs to live. With our Unified Edge we’re making it easier to power AI in the real world with flexible, secure systems that are simple to deploy, operate, and scale as demand grows.”

So what can customers expect from the new edge AI platform?

Under the hood of Cisco Unified Edge

Cisco Unified Edge is designed primarily for enterprises looking to run AI workloads and inferencing in a localized setting, removing the need for costly cloud-based infrastructure.

The platform can be scaled based on changing storage needs, which Cisco said will remove the need for “rip-and-replace upgrades”.

“The modular chassis offers CPU and GPU configurations, redundant power and cooling, high-performance SD-WAN networks, and pre-validated designs to support today’s applications and those yet to be imagined,” the company said in a statement.

Notably, Cisco aims to simplify integration of the platform, describing it as a “zero-touch deployment” system that can be easily managed through Cisco Intersight. This means considerations such as scaling and upgrades can be carried out without the need for “on-site specialist skills”.

ThousandEyes and Splunk integrations will also be available for Cisco Unified Edge, the company added, as well as built-in security features.

“Security is built-in at the device level and can extend to applying zero trust to every access, segmentation, and protection applications and AI models,” Cisco explained.

“This approach addresses the expanded attack surface at the edge, helping secure AI operations from physical and cyber threats.”

AI at the edge gains momentum

The announcement from Cisco marks the latest in a string of moves from big tech providers to bring AI closer to the edge, meaning data is processed closer to the source as opposed to relying on cloud infrastructure services.

There are a number of key advantages for businesses in this regard, according to Forrester, particularly with regard to latency and processing speed.

“Moving AI processing to edge devices enables rapid insights that traditional cloud-based setups can’t match,” Forrester VP and principal analyst Michele Goetz noted in a 2024 blog post.

“For sectors like healthcare and manufacturing, architects should prioritize proximity to offset latency. Low-latency, highly distributed architectures allow endpoints (e.g., internet-of-things sensors or local data centers) to make critical decisions autonomously.”

As ITPro noted in February 2025, hosting AI workloads at the edge also offers distinct advantages in terms of security.

In hosting data on localized devices, this essentially allows enterprises to keep a closer eye on storage and reduce exposure to unauthorized access.

MORE FROM ITPRO

• Cisco polishes its platform but the network is still king
• Cisco promises AI training for a million Americans
• 96% of businesses have low cyber-readiness, claims Cisco

As organizations reimagine how they connect people, applications, and data, one theme keeps resurfacing: agility. One of the most transformative ways this is being realized is via programmability and automation.

This article is the final in a three-part series, exploring the trends reshaping networking. In part one, we unpacked the rise of cloud-native networking, where service meshes, APIs, and open standards are reshaping connectivity.

Part one of this series explored the rise of cloud-native networking, where service meshes, APIs, and open standards are reshaping how enterprises build and scale distributed applications. In Part two, we examined secure by design principles, with a zero trust approach, secure access service edge (SASE), and quantum-safe planning becoming non-negotiable foundations for the next decade.

Automation is another pivotal strand to the change that’s taking place. Instead of relying on manual command-line interfaces (CLI), tomorrow’s networks will be defined by code, workflows, and application programming interfaces (APIs). From infrastructure as code (IaC) and observability to evolving NetOps skillsets, automation is not just about efficiency – it is becoming the DNA of modern networking.

Infrastructure as code is redefining how networks are built

Infrastructure as code has long been a staple in cloud computing but its adoption in networking is still maturing. Patricio Martelo, director of network solution architecture at Alcatel-Lucent Enterprise, says that while “IaC for networks is maturing at a rapid pace, it is not widely adopted outside of cloud environments”. He explains that networks don’t always have the same dynamic, on-demand nature as cloud workloads, so many organizations selectively adopt IaC in provisioning, compliance auditing, or remediation, but not yet in continuous operations.

The barriers are clear: skills gaps, legacy systems, and a lingering mindset shift. Hommer Zhao, director at Wiringo, emphasizes that trust is the biggest hurdle. “In the cloud, infrastructure as code is standard but in networking, many teams still rely on old systems and manual processes. The biggest hurdle is trust; people worry about giving too much control to code when downtime can cost millions,” he says.

Still, the risks of not adopting automation may be even greater. Sam Measures, head of domain Architecture at Markerstudy Insurance Services, explained that IaC unlocks agility when combined with guardrails such as pre-release code scanning. “The key to risk mitigation is to perform pre-release code scanning, deploy to non-production environments first, and leverage blue/green deployments,” he says.

The message is that enterprises cannot afford to treat networks differently from the rest of IT. As Zhao put it, agility without control is dangerous, but speed with safety nets like testing, rollback, and validation creates resilience, not fragility.

Observability is moving from visibility to intelligence

If automation is the engine of modern networking, observability is its navigation system. Traditional monitoring shows what is happening; modern observability aims to explain why – and increasingly, what to do next.

Martelo says observability must evolve to become advisory: “Tools must go beyond the current network status and help keep the network running optimally and securely both today and tomorrow.” He points to security scorecards, compliance benchmarks, and even automated remediation as the next frontier.

Zhao agrees, telling ITPro that most businesses today don’t just want raw data; they want actionable guidance. “Right now, most tools tell you what’s happening. What businesses really need are tools that suggest what to do about it,” he explains.

The role of real-time telemetry is also central here. As Measures explained, telemetry allows engineers to validate whether automated changes are still allowing traffic to flow as intended. But the challenge is the flood of raw data. “Telemetry is like a health check for the network. The challenge today is that we get too much raw data and not enough clear signals,” Zhao emphasizes.

To make telemetry useful, enterprises must invest in standardization. Formats like YANG and OpenTelemetry are gaining traction, helping unify fragmented data streams and turning noise into insight.

NetOps skills are shifting from CLI to coding

One of the most significant implications of programmability and automation is cultural. Traditional network engineers – long masters of the CLI – are being asked to think and work like software developers.

Martelo says resistance is still strong: “A large portion of network engineers still resist automation and only trust the CLI, which in many cases is due to fear of losing control.” His advice is pragmatic: start small. “The first step is learning how to code, preferably in Python. Once you understand one programming language, you can understand any piece of code or script if you look at it long enough.”

From there, engineers can build toolkits around Git, APIs, and orchestration platforms like Ansible or Terraform. Zhao puts it more simply: “Networking pros need to get comfortable with coding basics. You don’t need to be a full software developer, but knowing how to use APIs, automation scripts, and version control is becoming essential. It’s about thinking in workflows, not just commands.”

Measures emphasized that understanding DevOps methodologies and CI/CD delivery models is just as important as coding. “Network engineers need to focus on understanding the fundamental principles of DevOps methodologies. They need this foundation to properly embrace Infrastructure-as-Code,” he says.

Even certifications will need to adapt, as Zhao tells ITPro:

“Certifications will have to change. They’ll still cover the basics of how networks run, but they’ll also include automation, cloud integration, and coding. It’s less about memorizing commands and more about proving you can design and manage networks in a smarter way.”

The road ahead: balancing standards, ecosystems, and lock-in

As programmability becomes mainstream, the question remains: will the industry converge on open standards, or fragment into vendor-specific ecosystems?

“Customers who prefer a best-of-breed approach or want to protect themselves from vendor lock-in will demand open standards,” says Martelo who sees both forces at play. “But vendors will innovate above and beyond the open standards to create a differentiated value proposition.”

Measures, however, is more skeptical, suggesting that abstraction languages like Terraform may be the only way to achieve true interoperability across ecosystems. Without them, enterprises risk being pulled deeper into proprietary silos.

Zhao is blunt: “The hardest part is consistency. Each cloud provider has its own way of doing things, so when a company uses several clouds, it ends up juggling different tools and policies. Without common standards, automation risks becoming patchwork.”

The reality is that hybrid and multi-cloud networking will remain complex. Unifying observability and automation across diverse environments require trade-offs: either accept vendor lock-in for simplicity or pursue open standards with the added burden of managing complexity.

Programmable networks as the new baseline

Across this three-part series, we’ve seen a consistent theme: the future of networking is cloud-native, secure by design, and programmable. Cloud-native networking gave us agility, APIs, and service meshes. Secure networking established zero trust and SASE as baselines. And now, programmability and automation (Part 3) are redefining how networks are built, observed, and operated.

As Zhao explained, programmability is not about replacing humans with code. It is about giving humans the confidence, safety nets, and intelligence to run networks at the speed of business. Or as Martelo tells ITPro, the future is one where “code, not command lines, defines how infrastructure is built and managed.”

For business leaders, the implications are profound. Networks are no longer static infrastructure; they are dynamic, software-defined systems that demand new tools, new mindsets, and new skillsets. The organizations that embrace Infrastructure-as-Code, invest in observability, and retrain their NetOps teams for automation will not just survive the next decade—they will thrive.

Network design is being transformed faster than ever, as firms move to meet workers in new environments, apply simplified security on top of complex infrastructure and embrace cloud-native work.

This article is part two of a three-part series, exploring the trends reshaping networking. In part one, we unpacked the rise of cloud-native networking, where service meshes, APIs, and open standards are reshaping connectivity.

However, agility alone is not enough. As threats multiply and data spreads across clouds, branches, and home offices, networks must be secure by design. This means building in zero trust principles, embracing frameworks like secure access service edge (SASE) and security service edge (SSE) , as well as preparing for a quantum-safe future.

In this article, we consider how organizations can navigate these shifts, the barriers they must overcome, and the opportunities that come from treating security as a foundation rather than an afterthought.

Zero trust must be the baseline

The idea that a strong perimeter can protect networks is no longer credible. Remote work, SaaS adoption, and multi-cloud strategies mean attackers can exploit weaknesses far beyond the corporate firewall. Zero trust flips the model, requiring continuous verification of every user, device, and workload.

“Zero trust isn’t a buzzword, it’s a survival strategy,” says Axel Maisonneuve of the BSV Association. He cites the Colonial Pipeline ransomware attack as a stark reminder that a single weak VPN login can bring down critical infrastructure. “The biggest barriers are not technical, but cultural: moving from ‘inside equals safe’ to ‘every access must be verified, always.’ That shift requires leadership buy-in and a willingness to rethink legacy assumptions about trust.”

Kevin Curran, IEEE senior member and professor of cybersecurity at Ulster University, tells ITPro that while zero trust is realistic, organizations “typically struggle most with identity management, followed by network segmentation”.

“Establishing a robust, centralized IAM is challenging,” he says. “Weak identity governance leads to vulnerabilities that an attacker could exploit.”

Alan Stewart-Brown, VP EMEA at Opengear, adds that resistance to change remains the biggest brake: “Teams often favor VLANs and firewalls and underinvest in identity, segmentation, and continuous verification. The most difficult challenge is making identity and access operational at scale, with policies, health checks, step-up authentication, and role lifecycle management across branches, cloud, and data centers.”

The consensus is clear: zero trust is achievable, but it requires both cultural change and investment in automation. Identity must become the anchor point for security in distributed networks.

SASE and SSE unify access and security

If zero trust is the principle, then SASE and SSE are the architectures that put it into action. These cloud-delivered frameworks combine secure access with networking, ensuring consistent policy enforcement no matter where users connect from.

“Pre-pandemic, a global bank might have had 200 branch offices, each with its own firewall and proxy. Security was inconsistent, depending on where you logged in,” BSV Association’s Maisonneuve explains. “SASE and SSE unify that mess. They enforce the same security policies whether you’re at HQ in London, at home in Lisbon, or in a café in São Paulo.”

Greg Keller, CTO at JumpCloud, tells ITPro that the shift to remote and hybrid work has necessitated an overhaul of the security environment, particularly where consistent identity verification and threat detection is needed.

“As organizations shift away from traditional office environments, cloud-delivered security frameworks are becoming central to protecting distributed workforces and maintaining consistent policy enforcement,” Keller explains.

“With AI integrated into SASE, businesses benefit from smarter threat detection and automated policy management.”

But migration is not straightforward. Stewart-Brown notes that “when enterprises shift from SD-WAN to SASE/SSE, the first problem is policy sprawl. SD-WAN handles routing and QoS (Quality of Service), while cloud security platforms apply separate access and inspection rules. Managed in different consoles, these policies drift, conflict, and leave gaps.”

Nicholas DiCola, VP customers at Zero Networks, emphasizes that application support is another hurdle: “Routing everything through SASE will require testing of applications to ensure they don’t impact the users.”

For enterprises, the key is balancing resilience with user experience. As Maisonneuve warns, if security degrades productivity, employees will find ways to bypass it. The most successful migrations will unify oversight, automate policy enforcement, and focus relentlessly on usability.

Quantum-safe networking is coming

While zero trust and SASE address today’s threats, quantum computing raises questions about tomorrow. Once practical quantum machines emerge, widely used encryption methods like RSA and ECC could be broken, leaving decades of sensitive data vulnerable.

“The urgency is real because attackers are already collecting encrypted data today, intending to decrypt it once quantum computers are powerful enough,” explains Michael Murphy, deputy CTO at Arqit. “This puts industries that handle sensitive, long-lived data – like telecoms, financial services, healthcare and government – at greatest risk.”

Curran agrees, stressing that banking, payments, and national security should move first. But he frames the challenge as medium-term: “Quantum computers capable of breaking current encryption are not yet practical, but the risk of ‘harvest now, decrypt later’ attacks make preparation urgent.”

So, how should enterprises start? Stewart-Brown says the near-term priority is "crypto-agility: the ability to adapt cryptographic algorithms, keys, and protocols quickly and safely when risks, standards, or vendors change”. That begins with an inventory of where cryptography is used and who owns it.

Pragmatism is the watchword. Curran recommends piloting new algorithms such as NIST’s CRYSTALS-Kyber in low-risk environments. Murphy emphasizes symmetric key techniques as an immediate resilience boost.

“Most of the work will be in the certificates used for encryption at rest and in transit,” adds DiCola. “Creating a plan to do that starting now is the most pragmatic path.”

The takeaway is not to rip and replace, but to prepare. Enterprises should adopt hybrid approaches, testing quantum-safe methods alongside existing standards, while keeping an eye on emerging NIST frameworks. Early movers will not just mitigate risk, they will demonstrate resilience to customers and regulators alike.

Toward convergence: security as part of the continuum

Looking ahead, experts suggest that zero trust, SASE/SSE, and quantum-safe technologies will increasingly converge. Curran predicts that as firms adopt more AI policies, they are likely to unify this with zero trust verification, quantum key management, and SASE connectivity.

Murphy adds that “in five years these approaches will be seen less as separate tracks and more as parts of the same resilient framework.”

Maisonneuve frames the goal as “invisible security” – controls that work in the background, seamlessly protecting without slowing the business down. This perspective ties directly back to part one of this series, where cloud-native networking was shown to require not just technology, but cultural change. Security by design represents the next phase of that evolution: it is about making protection native, continuous, and adaptable, rather than bolted on after the fact.

It also sets the stage for part three: programmability and automation. If zero trust defines the principle and SASE/SSE provides the architecture, programmability is the engine that will make them scalable. Infrastructure-as-Code, policy-as-code, and advanced observability will allow enterprises to automate identity checks, enforce segmentation, and embed quantum-safe cryptography into daily operations. As networks become more software-defined, the distinction between security and operations will blur, requiring NetOps teams to balance traditional skills with automation and code-driven agility.

Secure by design networking is no longer optional. As Stewart-Brown emphasizes, relying on static controls like VLANs and firewalls leaves enterprises vulnerable to lateral movement and blind spots. The only sustainable strategy is to embed zero trust, adopt cloud-native frameworks like SASE and SSE, and prepare for a quantum-safe future.

The path forward requires cultural change, investment in identity and automation, and a commitment to resilience over convenience. But the reward is a network that doesn’t just connect the enterprise—it protects it, invisibly and continuously, in a world of constant change.

In the final part of this series, we will explore how programmability and automation will take these principles further, ushering in an era where networks are defined by code, enforced by policy, and continuously observable. The future of networking is hybrid, programmable, and secure by design – and the time to prepare is now.

The way networks are built, secured, and managed is undergoing a profound shift. As organizations embrace distributed applications, remote work, and increasingly complex IT environments, networks must evolve to be more cloud-native, more secure by design, and more programmable than ever before.

This article is part of a three-part series exploring the trends reshaping networking and the practical implications for businesses preparing for the decade ahead.

In part one, we examine the rise of cloud-native networking, from service meshes and application programming interface (API) driven control to the challenges of integrating across multi-cloud and hybrid environments and shift to open standards.

Cloud-native networking isn’t just a buzzword—it represents a fundamental redesign of how enterprises think about connectivity. Instead of being anchored in static infrastructure, networks are increasingly application-centric, designed to deliver agility, scalability, and resilience.

“Service meshes have fundamentally shifted network design from infrastructure-centric to application-centric models allowing a separation of data and control,” explained Marie-Claire Dwek, CEO at Newmark Security. This means that routing, security, and observability can be managed at the application layer, rather than buried deep in the infrastructure stack.

At the same time, APIs are transforming how networks are configured and consumed. Jean-Philippe Avelange, CIO at Expereo, explains to ITPro: “Network APIs have moved from being bolt-on conveniences to foundational building blocks. By exposing standardized, programmable interfaces, they allow developers to embed connectivity directly into CI/CD pipelines and enterprise applications.”

This shift is not just technical, as it changes who holds the keys to the network.

“Rather than pushing networking decisions down to the platform or infra teams, we’re seeing control planes shift upward to developers and security architects,” says Mayur Upadhyaya, CEO at APIContext. “This is not just about agility; it’s about policy enforcement at the point of orchestration.”

For enterprises, the benefit is clear: service meshes and APIs allow them to scale distributed applications, enforce consistent policies, and reduce human bottlenecks. But the transition is not without challenges.

Multi-cloud and hybrid networking bring complexity

While cloud-native principles promise agility, they also introduce new forms of complexity—particularly when enterprises operate across multiple cloud providers or hybrid environments.

“The most pressing integration challenges in multi-cloud and hybrid environments stem from the fundamental differences in how cloud providers and on premises infrastructure handle networking, security, and operational models,” emphasizes Dwek.

Each provider comes with its own identity systems, routing methods, and security models. Trying to stitch them together can create inconsistencies, blind spots, and extra costs.

Upadhyaya agrees but takes the view that the challenge is as much about data as it is about the infrastructure itself. “The biggest challenge isn’t the plumbing, it’s the metadata. Each cloud exposes identity, policy, routing, and telemetry in slightly different ways. This leads to gaps in trust, context, and visibility,” he says.

Enterprises are learning this lesson the hard way. Avelange points out that what works in one environment often breaks in another: “Policy enforcement is uneven, observability is fragmented, and resilience depends on manual workarounds.” IDC’s 2025 survey, commissioned by Expereo, revealed that over half of enterprises experienced revenue loss tied directly to network failures in multi-cloud scenarios.

The message is clear: multi-cloud is not a silver bullet. Without investment in abstraction layers, open control planes, and continuous compliance monitoring, it can become a liability rather than a source of flexibility.

Vendor consolidation and the role of open standards

Another trend shaping cloud-native networking is vendor consolidation. Cisco, Juniper Networks, VMware, HashiCorp, and others are snapping up companies and rolling out broader suites of tools. The question is: does this make life easier for enterprises, or does it lock them into bigger silos?

“The promise is interoperability, but the reality often looks like branded silos with slightly better integration,” says Upadhyaya. He cautions that unless customer pressure or open standards force it, consolidation often prioritizes revenue over user flexibility.

Avelange takes a more balanced view: “On one hand, mergers often deliver tighter integration inside a single vendor’s portfolio. On the other hand, consolidation also risks creating even larger proprietary silos. The deciding factor will be standards adoption.”

Open standards are therefore essential to keep the networking ecosystem fair, flexible, and future-proof. Initiatives like the Kubernetes Gateway API, SPIFFE (Secure Production Identity Framework For Everyone) for workload identities, and eBPF (Extended Berkeley Packet Filter) for kernel-level observability are helping to reduce vendor lock-in and enable portability.

“Open standards are the antidote to lock-in,” Avelange tells ITPro. “They provide enterprises with common building blocks that work across vendors and environments, making it harder for any single provider to trap customers inside proprietary models.”

Dwek also warns that under-standardization remains a risk, particularly around cross-cluster service discovery and policy enforcement. Without agreed formats and governance models, enterprises may find that their cloud-native environments become as brittle and fragmented as the legacy systems they were supposed to replace.

Preparing for a hybrid, programmable future

If there’s one consistent theme across all expert perspectives, it’s that the future of networking will be hybrid and programmable by design. Enterprises will continue to rely on cloud provider-native services for core connectivity, but they’ll increasingly layer open, programmable control planes on top to achieve flexibility and policy consistency.

Avelange says the optimal position over the next five years is one of a hybrid balance.

“Enterprises should lean on cloud provider-native networking services for what they do best: baseline connectivity, global reach, and built-in scalability. At the same time, relying solely on native features risks fragmentation and lock-in. The real differentiator will come from open, programmable control planes layered on top.”

Dwek also emphasises that success in cloud-native networking depends as much on people as on technology: “The human element – training teams and evolving organizational practices – often determines success more than technical architecture choices. Organizations should invest as much in people and processes as they do in technology platforms.”

The cultural shift cannot be understated. IT teams must evolve from configuring routers to writing code, from managing tickets to embedding networking into DevOps pipelines. Upadhyaya adds that this requires “new skill sets: not just YAML fluency, but also an understanding of developer ergonomics, observability tooling, and policy-as-code frameworks”.

For business leaders, the takeaway is that cloud-native networking is not optional. It is the baseline for supporting AI workloads, enabling global operations, and delivering seamless digital experiences. The organizations that embrace it early – while balancing standards, skills, and strategy – will be best positioned to thrive.

Coming up in this series

This article has explored the first piece of the networking future: cloud-native principles, service meshes, APIs, and standards. But this is only the start.

In part two, we turn to security, where zero trust principles are becoming the baseline for distributed networks. We’ll also explore the growing adoption of secure access service edge (SASE) and security service edge (SSE) architectures to support remote and branch access, as well as how organizations are beginning to plan for a quantum-safe future.

Finally, in part three, we look at how networks are becoming programmable and automated. From infrastructure as code (IaC) to observability for traffic flows and policies, we’ll consider how NetOps teams are evolving – shifting from traditional CLI-based management to code-driven automation.

Together, these three articles highlight not just where networking is heading, but what businesses need to do today to prepare for a future where agility, resilience, and programmability define success.

The flow of data is core to businesses all around the world – how much can be transmitted, and at what speed, is a fundamental aspect of IT systems ranging from cloud computing to the frontier of AI models.

In data centers, the tech sector is focused on reducing latency, improving bandwidth, and reducing energy consumption to improve AI performance and lower operating costs. But achieving these goals by shifting about current technology can only go so far.

Increasingly, firms are looking to cutting-edge technologies and approach such as all-photonics networks, which use photons for all processes including data transfer and storage, rather than electronic components. How close are we to achieving this technology? And how could it benefit every business.

In this episode, Rory is joined Chris Wright, senior vice president global engineering and CTO at Red Hat, to discuss the potential benefits of photonics based networks and how they could upend current network architecture.

Highlights

"[A]s we introduce all photonics networks and this end to end connectivity, extend it into the data center, we're reducing the latency such that it starts to feel much more like it's just another server and another rack in the same data center. We're literally approaching the speed of light in terms of how we transmit signals from end to end and the latency in, say, thousand of kilometers is still measured in a small number of milliseconds in a way that again, it starts to feel like you're very close."

"[T]o really inspire change you need to show something better, not just at a technical level but at some kind of an outcome. Why is this better for a user, for a business? And so while we have a really deep technology focus in the IOWN Global Forum, in order to bring that to life we've always had an interest in thinking about it through the lens of use cases. So the global forum's membership base isn't only the technology providers, the network builders. It includes members who represent different industry focus areas, and that's really been since the beginning."

"Also that low latency, high bandwidth connectivity gives you flexibility in where you actually place data centers. So you could place data centers in areas that are easier to power, easier to cool for example, maybe not in the core of an urban environment, maybe just slightly outside of a city closer to a water source for cooling or as a power source, or closer to to a wind farm or solar farm."

Footnotes

• What is latency?
• Is latency always important?
• The answer to data center sustainability? Industry rallies around light-based networking systems

Subscribe 

• Subscribe to The IT Pro Podcast on Apple Podcasts
• Subscribe to The IT Pro Podcast on Spotify
• Subscribe to the IT Pro newsletter
• Join us on LinkedIn

Russian government-linked hackers are exploiting unpatched Cisco networking devices to spy on critical infrastructure organizations.

The attackers are mainly targeting organizations in the telecommunications, higher education and manufacturing sectors, with known victims in a number of geographic regions, including North America, Asia, Africa and Europe.

The group carrying out the attacks is believed to be part of the Russian Federal Security Service's (FSB) Center 16, and has a number of names, including Static Tundra, Berserk Bear, Energetic Bear and Dragonfly.

It's exploiting Simple Network Management Protocol (SNMP) and end-of-life networking devices running a seven-year-old unpatched vulnerability in Cisco Smart Install (SMI).

"For years, Static Tundra has been compromising Cisco devices by exploiting a previously disclosed vulnerability in the Smart Install feature of Cisco IOS software and Cisco IOS XE software (CVE-2018-0171) that has been left unpatched, often after those devices are end-of-life," said Cisco Talos researchers Sara McBroom and Brandon White in an advisory.

"We assess that the purpose of this campaign is to compromise and extract device configuration information en masse, which can later be leveraged as needed based on then-current strategic goals and interests of the Russian government. This is demonstrated by Static Tundra’s adaptation and shifts in operational focus as Russia’s priorities have changed over time."

Over the last year, the group has been spotted collecting configuration files for thousands of networking devices associated with US organizations across critical infrastructure sectors.

On some vulnerable devices, it modified these configuration files to enable unauthorized access, through which they then carried out reconnaissance in their victims' networks - revealing their interest in protocols and applications commonly associated with industrial control systems.

The group has been compromising network devices for more than ten years, focusing particularly on devices accepting legacy unencrypted protocols like SMI and SNMP versions 1 and 2.

It's also made use of custom tools against certain Cisco devices, such as the malware known as SYNful Knock in 2015.

"Once they establish initial access to a network device, Static Tundra will pivot further into the target environment, compromising additional network devices and establishing channels for long-term persistence and information gathering," said McBroom and White.

"This is demonstrated by the group’s ability to maintain access in target environments for multiple years without being detected."

Cisco Talos is urging customers to apply the patch for CVE-2018-0171 or, if that's not an option, to disable Smart Install.

"The threat extends beyond Russia's operations — other state-sponsored actors are likely conducting similar network device compromise campaigns, making comprehensive patching and security hardening critical for all organizations," McBroom and White warn.

"Threat actors will continue to abuse devices which remain unpatched and have Smart Install enabled."

MORE FROM ITPRO

• The convergence of network and security – how it helps achieve business outcomes
• Russian hackers tried to lure diplomats with wine tasting
• ‘All US forces must now assume their networks are compromised’ after Salt Typhoon breach

Enterprise networking is at a pivotal moment. Faced with rising operational costs and increasing expectations from end users, value-added resellers (VARs) and managed service providers (MSPs) are squeezed between two extremes: complex, costly enterprise systems on one side and fragmented, unreliable, low-cost solutions on the other. In the middle lies an expanding opportunity, but only for those willing to rethink how success is achieved.

Margins are shrinking not because of competition but because of complexity. Multi-tiered licensing, inconsistent support, and one-size-fits-all architectures result in repeated visits to customers, longer deployment times, and higher overheads. For partners aiming to build profitable service businesses aimed at enterprise customers below the FTSE 100, complexity is no longer just inconvenient; it’s a direct obstacle to growth.

What’s emerging now is a move toward appropriately sized solutions. Enterprise-grade reliability no longer requires an enterprise-sized IT team. The best platforms today are designed for businesses needing robust cloud connectivity, seamless remote management, embedded security, and AV/IT integration, without excessive overheads or unnecessary features. Solutions that emphasise simplicity, usability, and fit-for-purpose design are gaining popularity because they work. Not eventually, with the support of three site visits, escalations, and unnecessary costs being incurred, but right from day one.

A new breed of vendors is entering the market to deliver solutions at this level. They are not trying to be bigger and more complex than Aruba, Cisco, or Juniper.. Nor to be cheaper than the Chinese consumer-grade manufacturers. Their ambition is different: to be the easiest to do business with, offering a right-sized portfolio for the ‘Unfortunate 50,000’. This term was initially used 15 years ago to describe companies that lacked access to enterprise Wi-Fi technology, because it was too complex and expensive. But history repeats itself – and here we are again, with a demand versus supply gap. This time, however, there is a solution.

Small and medium-sized enterprises (SMEs) have the same needs as larger enterprises in terms of wired and wireless network performance, SD-WAN, redundancy, or SASE security. They just don't have the budget to spend millions with the biggest vendors.

This shift assigns new responsibilities to the new breed of vendors like ourselves. It’s no longer enough to sell a capable product. It must install smoothly, integrate easily, and be manageable without a week of training. That’s where professional services, once optional, are becoming vital. Pre-staging equipment, validating designs, assisting with deployments, and conducting post-install health checks or free site surveys are now critical in ensuring customer satisfaction and reducing long-term support costs.

Training must also adapt. The focus isn’t on certification for its own sake or LinkedIn points, but on operational readiness. Partners need tailored enablement that centers on real-world results: Wi-Fi 7 or AV-over-IP workshops, streamlined onboarding for new technicians, and tools that get them operational quickly, without sacrificing quality.

Ultimately, the future of the networking channel is NOT about features anymore; it is about enablement. Vendors must rapidly eliminate friction in pricing, licensing, support, and partner programs.

For partners, every moment spent chasing quotes or troubleshooting a misconfigured VLAN is time not spent growing their business. What partners need - now more than ever - are solutions that scale with their expertise, deliver dependable performance, and generate predictable profits.

Autonomous vehicle navigation, debit transactions, and security footage-focused action are all examples of where latency – the delay between an instruction to transfer data and the data being transferred – can wreak havoc.

Companies like WEKA have called latency “the performance killer in AI workloads” and Amazon Web Services (AWS) has identified low levels of latency as being as important as the input data for generative AI models. So, when is low latency vital and when is it a nice to have?

To begin with, it’s not as simple as an either/or. Elliot Marx, co-founder of data platform Chalk, says that there are two types of latency on which IT leaders should be focused.

“One is just how fast the servers can give you back an answer so that you can make a decision,” he explains. “That’s really important in a credit card authorization loop, like when you swipe your card at Costco and you want to say yes or no to a purchase, that needs to be a really low amount of time.

However Marx sees there as being a second form of data latency that can also have a significant impact.

“How old is your data? So you could serve data that you found out about a long time ago really fast, and you might call that low latency, but it still could be high latency in terms of, well, this is really stale data.”

While conversations about AI often fixate on the latest and greatest, Marx says it was FinTech that had to figure out the latency conundrum first, ten to fifteen years ago, because of the ever-present threat of fraud. Now, he’s seeing that same urge for latency reduction spilling out onto social media and e-commerce platforms, places where the quicker they can learn your behaviour, the quicker the company can shape it. He gives TikTok as an example of a social media giant that has optimized latency in order to further shape their industry-leading algorithm.

However, low latency isn’t just important in the digital world. Another example where its reduction is key is when physical security is involved, such as when a video feed is being fed into an automated process so that safety threats – whether they’re security issues like thieves or health and safety problems like tripping hazards – can be identified and addressed. Val Cook, chief software architect at Blaize, says that as the industry demands less off the shelf solutions and more bespoke choices, the importance of low latency is going to continue to grow.

“I would say 60 to 70 percent of the requests we get even for POCs [proof of concepts] with customers are just not simple cookbook solutions,” Cook tells ITPro. “They want to know if someone ducked down and they appear to have slid under a car, and they're cutting off the catalytic converter…So, we see that over and over and over again. Could you do this?”

On the customer or user side, Cook also thinks that low latency solutions are in a higher demand because it’s human nature to want a product—whether it’s a large language model (LLM) or a security system interface—to react quickly and to match our pace of work.

“Just the way humans work, the response time is going to[need to] be much more sensitive to being able to provide visual feedback and visual input.”

This is a key consideration when it comes to siting data centers. For example, a company looking to process its data in regions with plentiful renewable energy and therefore green data centers, such as Iceland, will also have to contend with increased latency.

The exact amount will depend on where one’s data was originally stored and processed. In the case of Shearwater Geoservices, which embraced Icelandic data centers in 2024 for reduced operating costs, the firm saw its latency increase from 3-4 milliseconds in the UK to 20-30 milliseconds in Iceland. This was ultimately deemed acceptable for real-time operations.

Workers on the move, such as those in field service management, also have to factor the latency of 5G networks or satellite broadband into their operations. This has to be weighed against what one considers ‘acceptable’ latency depending on the task at hand, and generally compares unfavorably to full-fiber broadband.

For example, in the UK the average latency over Wi-Fi is 19 milliseconds, per Uswitch data, compared to 33 milliseconds over 5G and 42 milliseconds over 4G. A recent Ookla report on Starlink satellite broadband found UK latency was 41 milliseconds on average.

How do organizations reduce latency?

Tech giants are aware of the need to reduce latency. OpenAI, for example, provides an entire guide to reducing latency in AI implementations, one which includes a paradoxical piece of advice: don’t automatically assume that an LLM is your best option.

The firm, along with all major AI model developers, offers a range of models designed to meet the latency needs of users. For example, its lightweight GPT-5 nano model is less sophisticated but far faster to produce outputs than GPT-5, while GPT-4o mini Realtime has such low latency that it can be used for real-time audio inputs.

Other models, such as Gemini 2.5 Flash, Mistral Small, and Nvidia’s Llama 3.1 Nemotron 70B also boast very low latency.

For Marx, his team has found that simplifying the programming language they’re using has helped to drastically reduce latency.

“What we found is that you have to write really low level systems code. And so we end up writing a lot of Rust and a lot of C++,” he tells ITPro.

“Those aren't really the languages of data science and of machine learning engineering, though…What we do is we treat Python and SQL kind of like a domain-specific language, and we reinterpret it and run it as much as we can in lower level systems languages that are much, much, much faster, like orders of magnitude faster.”

Between simplifying the language being used and avoiding reverse extract, transform, load (ETL) in favour of computing on the fly, Marx says his team is able to bring complex functions down from a ten to fifteen second wait time down to 15 milliseconds.

For Cook, the key – instead of using a traditional batch method – is to build a system where multiple tasks can happen dynamically, meaning an entire sequence doesn’t have to happen all at once. This asynchronous approach reduces the amount that latency compounds.

“When I have to declare my graph, run a detector, run a face recognizer, run an action recognizer, and so forth, when I have to declare those all in a pipeline and then run them and then see what you get at the end, you get two challenges,” he says.

“One is you increase your latency dramatically. And number two, to your point, you waste compute.”

Still, there are examples where lower latency isn’t as high of a priority. Marx points to scenarios where business leaders need to be able to see a dashboard that assesses the health of their company – data that’s unlikely to be needed up to the millisecond.

“They could still be complex pipelines but you might not care so much about them being really up to date. I think one of the trends we're seeing in the industry though, is that people want to have one source of truth of their data and not be implementing that same sort of complex logic in a bunch of different places, because inevitably you end up writing it a little bit differently in each spot.”

This is why he sees computing on the fly as the approach, saying that reverse ETL “is going to die” as people prioritize creating systems that don’t require moving around and duplicating data as much.

In the name of efficiency, whether it’s building pipelines and processes or writing in simpler languages, the industries using tools like generative AI will continue to prioritize latency. It lessens compute cost, increases productivity and marketability, and creates opportunities for systems to run faster. In a world where the money is in the milliseconds, latency will be front of mind for industry experts as this portion of the industry evolves.

A new business-to-business telecoms firm has launched, formed through the merger of Virgin Media O2 Business and Daisy Group.

The new company, O2 Daisy, will have a large customer base - 45.8 million broadband, mobile, phone and home subscribers - and scaled fixed and mobile networks of Virgin Media O2.

This will be combined with the platforms, product sets, and business-to-business experience of Daisy Group.

The company will be led by Matthew Riley as chairman and Jo Bertram as CEO.

“Today we start on our mission to shake up the market and provide the technology needed to make every business better. With the launch of our new standalone company, we’re putting our focus on what really matters - helping businesses of all sizes connect and grow," said Bertram.

"By bringing together two experienced teams and supercharging them with scaled networks, agile systems and comprehensive products, we’re ready to deliver exceptional outcomes for our customers.”

O2 Daisy will target businesses from Small Office Home Office (SOHO) outfits to large enterprises and public sector organizations. It will, right from the start, be the UK’s second-largest small and medium-sized business (SMB) solutions provider.

Teams have been merged by the two firms, with O2 Daisy now having bases across the country, including hubs in Nelson, Sheffield, Manchester, London, Reading, and Bournemouth.

What to expect from O2 Daisy

The company will start out with around £1.4 billion in annual pro forma revenues and earnings of £150 million, with a customer base spanning hundreds of thousands of SMBs, large enterprises, public sector bodies, and indirect partners.

From today, it will offer a broad product portfolio that includes data and Wi-Fi solutions, workplace tools, 5G private networks, cloud-based services, cybersecurity offerings, and O2 Motion movement insights.

This includes an expanded range of products, such as cloud-based products and Teams Phone Mobile, as well as other on-net connectivity services. More solutions are expected to be released in the coming months, according to Virgin Media O2..

Similarly, the new firm will continue to use its existing brands during the integration phase, with further brand and operational updates expected in the coming months. For the time being, there’s no change to any services and all help and support contacts remain the same.

“This is an exciting day for O2 Daisy, as we officially launch and bring together two highly successful companies into one powerhouse team. In a fast-changing world, growth is inextricably linked to the ability to access world-class IT and communications infrastructure," said Riley.

"Driven by the entrepreneurial spirit we’re known for, O2 Daisy will revolutionize the telecommunications and IT landscape, by creating connectivity offerings fit for UK businesses of all sizes – underpinning their own growth stories.”

MORE FROM ITPRO

• How Virgin Media O2 saved “millions” by enhancing data center efficiency
• Virgin Media O2 Business overhauls its approach to partner development
• Virgin Media O2 cuts data center cooling costs by £1 million per year

Fast and reliable connectivity has always mattered when it comes to innovation, but the widespread adoption of remote and hybrid work as well as increasing reliance on cloud-based technology such as generative AI means that the need to deploy fiber-to-the-premises (FTTP) has never been more pressing.

FTTP is where fiber optic cables carry data at high speed directly to buildings. This is different to fiber-to-the-cabinet (FTTC), where fiber optic cables run into a street cabinet, which is connected to buildings via an existing copper cable network.

FTTC is the most common type of broadband but is imperfect, as the use of copper cables in the last leg means connections are likely to experience electrical interference. This can cause delays in data transmission. On the other hand, FTTP offers a superior and more reliable connection with lower latency. This is why FTTP is ideal for powering the AI revolution.

“AI applications, including legacy machine learning, data analytics, and Internet of Things devices, require substantial bandwidth and low latency. This surge in demand is prompting fiber broadband providers to upgrade their networks to handle massive data transfers efficiently,” wrote the Fiber Broadband Association in a report published in March.

Despite this great need, the US is often accused of lagging far behind international partners on fiber availability.

Iceland is the top-ranked country for fiber penetration, excluding FTTC, with 90.9% of its population accessing their broadband via the connection as of June 2024 per OECD data. Korea’s 90% penetration rate puts it third, while Spain is third with a rate of 88.3%. Looking closer at the OECD data, the US (24.5%) has one of the poorest penetrations. Only the Czech Republic (22.4%), Germany (12.2%), Austria (12%), Greece (!1%) and Belgium (9.3%) were ranked lower, while the OECD average was 44.6%.

The geography of the US

Full-fiber rollouts in the US may lag other developed countries, but looks can be deceiving.

“The narrative that US fiber is lagging behind others is a simplification. It fails to consider the fundamental realities of the American internet connectivity landscape,” says Rick Fulwiler, senior director of product management at NetScout

“From a geographical perspective, the vast size and diverse terrain of the US presents both logistical challenges and regional disparities in fiber optic internet infrastructure deployment,” adds Fulwiler. For comparison, rolling out fiber in South Korea is more manageable because it’s roughly the size of the state of Indiana.

Michael Misrahi, EY Americas’ telecommunications strategy and transformation leader, says that penetration in the US may seem low, yet this is down to the “large, rural skewed land mass” that doesn’t yet have the infrastructure in place for fiber connections to be deployed.

Cost issues hamper nationwide fiber plans

According to research by the Dell’Oro Group, fiber subscribers are forecast to account for a third of all broadband subscriptions by 2029. Cable’s market share is expected to fall from 62% to 53%, while the number of fixed wireless access (FWA), which relies on 5G technology, is expected to ramp up.

Misrahi expects an industry-wide push for more fiber connections across the US, especially in greenfield builds, but also in overbuilds where fiber will be built over existing copper lines. Even FWA connections that are being rolled out could eventually be replaced by full-fiber when and where economically viable, he says.

The cost of rolling out full-fiber infrastructure, especially in rural parts of the US, is a huge hurdle. “You can light up every apartment in a Seoul tower for about what it costs to trench half a mile of West Texas ranch land,” says Tyler Cooper, editor-in-chief of BroadbandNow Research, the analytics arm of data aggregation company BroadbandNow.

The $42.45bn US Broadband Equity, Access, and Deployment (BEAD) program is aimed at connecting everyone in the country to affordable, high-speed broadband. It was established during the Biden era with a fiber-first focus, but in early June, the Trump administration announced drastic changes to BEAD that will now take a tech-neutral approach.

“Unless there’s a u-turn on the fiber imperative, the US will keep slowly adding fiber miles every year while continuing to look up at our smaller, denser peers,” warns Cooper.

LEOs fill the connectivity gaps

A technology that Trump’s administration is likely to favor over fiber is low-Earth orbit (LEO) satellites developed by SpaceX subsidiary Starlink and Amazon’s Project Kuiper.

LEO satellites are already filling connectivity gaps, including in rural Vermont. Nevertheless, the state’s broadband leaders still believe that fiber should be a priority.

“LEO satellite services can help keep remote sites online, but finite capacity and lopsided uploads mean you’ll be hard-pressed to find a business that considers it a replacement for the gold standard [of fiber],” argues Cooper.

Misrahi adds that LEO satellites can experience spectrum interference and reliability issues, which can make the cost of relying on the technology less attractive. Ultimately, the best role for LEO satellites is likely to be as a bridge to other broadband connections, such as while rural communities wait for full-fiber networks to be deployed.

What is clear is that none of fiber, FWA or LEO satellites are a one-size-fits-all solution. “To connect its vast and diverse population, the US is going to require a multi-technology strategy, tailored to specific local conditions,” concludes Fulwiler.

The US Justice Department (DOJ) has settled a lawsuit contesting HPE’s $14 billion acquisition of Juniper Networks.

As part of the settlement, HPE will be required to make concessions to push the deal through. This includes divesting its ‘Instant On’ wireless networking division within 180 days, court filings show.

That means the company will be required to offload all assets, IP, and customers related with this particular area of the business.

Elsewhere, HPE is also required to hold an auction to license source code for Juniper’s AI Ops for Mist platform.

“The license will be perpetual, non-exclusive, and include optional transitional support and personnel transfers to facilitate competition,” the DOJ said in a statement on Friday.

While the settlement requires approval by a judge, this does pave the way for the acquisition, according to HPE chief executive Antonio Neri.

In a statement issued last week, Neri welcomed the move and said the settlement “preserves the intended benefits of this deal” and will create “greater competition in the global networking market”.

“For the first time, customers will now have a modern network architecture alternative that can best support the demands of AI workloads,” he said.

“The combination of HPE Aruba Networking and Juniper Networks will provide customers with a comprehensive portfolio of secure, AI-native networking solutions, and accelerate HPE’s ability to grow in the AI data center, service provider and cloud segments.”

HPE has had to battle for the Juniper deal

HPE first announced plans to acquire Juniper Networks in 2024, but the deal quickly sparked competition concerns. The company noted that the acquisition would create a new contender in the global networking market and enable it to double the scale of its networking business.

However, the DOJ announced legal action in January in a bid to block the deal, arguing that it would stifle competition in the space. If approved, the acquisition would mean Cisco and HPE would control more than 70% of the US networking market combined.

The DOJ specifically highlighted competition in the WLAN supply market as a key concern, suggesting that cutting competition in the space would harm innovation and impact customers.

HPE and Juniper were the second and third largest suppliers of WLAN equipment in the US, ranked behind Cisco.

A back and forth between the DOJ and HPE quickly unfolded in the wake of the legal action, with the latter suggesting the department was “divorced from reality” and that any block would benefit competitors.

The DOJ’s attempt to contest the deal also came after European and UK regulators approved the deal in August of last year.

MORE FROM ITPRO

• HPE just announced huge changes to its channel programs
• HPE GreenLake gets the agentic AI treatment
• HPE unveils Mod Pod AI ‘data center-in-a-box’ at Nvidia GTC

In early 2027, the UK’s analog phone service will be switched off. This will mark the end of a technology that has operated, without interruption, since 1879.

The end of the analog phone system has been on the cards for some time. Increasingly, analog voice, also known as the public switched telephone network or PSTN, has given way to internet-based services: voice over internet protocol (VoIP). The legacy digital ISDN network will be retired at the same time as the PSTN.

The move to replace analog lines with VoIP was first announced in 2017. And originally, the PSTN switch-off was meant to happen this year. That date was pushed back to allow telecoms operators more time to migrate their networks and to ensure provisions are put in place for vulnerable users of the consumer network.

Nonetheless, according to data from the Department of Science, Innovation and Technology, over two-thirds of landlines had moved to VoIP by May 2025. BT Openreach stopped selling new, copper-wire based lines in September 2023, so if a business needs phone services now, it will be VoIP.

The copper network will, though, live on for a while yet. Internet services running over copper lines, also known as fiber to the cabinet (FTTC) will carry on, until all addresses have access to faster, fiber to the premises (FTTP) services. There is no fixed date for this, but most telecoms providers expect it to be around 2030.

Ofcom, the telecoms regulator, points out that the decision to move away from analog lines was a commercial one by the service providers. But there is no doubt that the PSTN is now old technology, lacking the features and flexibility of VoIP.

And the PSTN is no longer reliable. According to Ofcom’s latest Connected Nations report, 2024 saw a 45% increase in PSTN “incidents”. Providers, for their part, have been forced to recycle or reuse old equipment for spares as much of the PSTN hardware is no longer made.

Network headaches for CIOs

There is little doubt, then, that it is better to invest in more modern technology, such as VoIP and fiber connections.

The issue is it is not just voice calls that are affected, or solely consumers.

There are millions of business phone lines that have not yet been migrated, across both analog and ISDN.

Some of these will be in small businesses or branch offices; remote locations or harder to reach facilities and locations such as small retail or hospitality outlets, workshops or warehouses that have little need for the extra features provided by VoIP. Phone systems are one of those technologies that just keep on going: why upgrade them if they are not broken?

Then there are the devices that use analog or dedicated ISDN circuits, for non-voice applications.

As the trade body TechUK notes: “Landlines aren't the only things that rely on the phone network. It also supports everything from healthcare devices, burglar alarms, and personal alarms, to home monitoring systems ATMs, card machines, traffic lights, motorway signs and railway signals.”

For CIOs this poses a number of challenges.

Firstly, companies need to identify where legacy landline and ISDN connections are still being used. Then they need to plan to migrate them to alternatives. These could be VoIP, but for some applications other options, such as cellular data, might be more appropriate or economical.

And, if a VoIP connection is possible, IT teams need to ensure their hardware is compatible with the proposed VoIP service. Not all provide the same levels of compatibility for non voice services.

If this seems a lot to unpack, it is. And there is not much time to do it. Awareness of the switch-off is still limited, especially among smaller businesses.

“Despite significant strides in the migration efforts, there is still a substantial backlog of PSTN lines awaiting transition,” Ben Avery, head of channel at telecoms and broadband provider Gradwell, tells ITPro.

“With over a million lines that still need to be migrated, it’s not just about telephony. Business systems including fire alarm systems and chip-and-PIN terminals to door-entry controls will be impacted by the switch-off,” Avery explains. “Within that figure, telecare stands out as a vertical that currently lacks readiness. Telecare is critical for elderly and vulnerable users and is a glaring example of where there is still plenty of work to do ahead of the 2027 deadline.”

3G gone, 2G going…

The situation, though, is made rather harder by the decision also to switch off 3G, and eventually, 2G mobile networks.

The 3G network has largely already shut down, with Vodafone and EE completing the switch-off in early 2024. Three has shut down most of its 3G signals, and O2 will do so this year.

The immediate impact of the 3G switch-off for businesses has been fairly limited; not many employees will still have phones that do not support 4G. The issue, as with landlines, are those hidden devices: signalling, sensors, alarms and IoT hardware.

Some of this equipment was designed before 4G, and cannot be upgraded. Hardware that uses SMS for signalling will still work on the 2G networks, as will low data rate devices. But SMS, or even GPRS (also known as 2.5G) is a poor substitute for even 3G connectivity. And 2G itself is due to switch off fully by 2033, with some providers completing the transition before then.

Ofcom's report estimates that 2.1 million devices still used 3G in 2024; the figure should have fallen since. But, although half of the legacy devices were consumer handsets, this still leaves a million or more business connections.

Then there are telecare devices; eCall systems in vehicles which make automatic calls in the event of an accident, usually over 2G; and some 7 smart electricity meters which, according to the House of Commons Public Accounts Committee, are still using 2G and 3G. This is aside from any specialist hardware businesses run.

It might not be a glamorous project, but IT teams would do well to hunt down those legacy systems and connections, landline and mobile, as soon as they can.

No one wants to learn that a critical service has stopped working because users, or worse still customers, start calling the helpdesk.

As businesses increasingly rely on fast, stable internet connectivity to operate, the limitations of terrestrial broadband infrastructure – especially in rural and remote areas – are becoming harder to ignore. Satellite broadband, historically considered a last resort due to high latency and cost, is undergoing a renaissance.

Satellite broadband has long been synonymous with rural communities: slow, expensive, and reserved for the most remote users. However, the advent of low Earth orbit (LEO) satellites, spearheaded by companies like Starlink, OneWeb, and Amazon’s upcoming Kuiper network, is changing that perception.

Promising lower latency, higher speeds, and global reach, these next-generation satellite networks have sparked renewed interest from enterprise users and governments alike. However, does satellite broadband truly have a viable business case in the UK and EU, or is it still a niche solution? The answer, as it turns out, depends on the use case, geography, and expectations.

Until recently, satellite internet was hindered by high latency and inconsistent service. Geostationary (GEO) satellites orbit at about 36,000 km, making real-time communication difficult. LEO satellites, by contrast, orbit at altitudes between 500 and 2,000 km, significantly reducing latency and, therefore, data transmission speeds. According to ABI Research, LEO satellites will expand from the current 7,000 to an astonishing 42,600 by 2032.

Simon Dumbleton, UK CTO at World Wide Technology, explains that the technical capabilities of LEO networks are altering the way satellite broadband is perceived. "Users have been generally reporting consistent and reliable performance, both in terms of download and upload speeds, as well as latency," he says. "The ground dish is super sophisticated, and it continuously tracks the satellites to maximize performance."

Yet, while performance metrics are improving, not all networks are created equal. Dumbleton notes that mobile broadband services via satellite are still in their infancy, with current capabilities limited primarily to messaging and emergency services. However, the Non-Terrestrial Networks (NTN) standards evolving under 3GPP are paving the way for richer applications over satellite networks in the near future.

The current state of satellite broadband offerings

The new wave of satellite broadband providers is fundamentally different from legacy systems. LEO constellations offer lower latency and higher bandwidth, rivaling traditional broadband in performance.

Starlink

Operated by SpaceX, it has emerged as the most mature offering. As of mid-2025, it has deployed over 7,000 satellites with plans to scale to more than 42,000.

The company offers speeds of 25–220 Mbits/sec with latency as low as 25ms. Starlink’s standard business package in the UK costs start from £39 per month for 50GB of data up to £324 per month for 1TB of data, plus an upfront hardware cost of £299, supporting priority data and multiple IP addresses.

OneWeb

The UK-headquartered firm, partly owned by the British government and Bharti Global, completed its first constellation phase with 648 satellites. While OneWeb focuses more on enterprise and governmental clients than consumers, it is also targeting in-flight connectivity and maritime markets.

Project Kuiper

Amazon plans to launch 3,236 LEO satellites, with services expected to begin in 2025. Though it lags behind Starlink and OneWeb in deployment, Amazon is betting on its cloud integration to add business value.

Across all three providers, the promise is similar: broadband without boundaries, capable of supporting remote operations, disaster recovery, and mission-critical IoT infrastructure.

Niche but growing demand

The adoption of satellite broadband remains limited in Europe but is gaining momentum. According to Ofcom, as of late 2024, around 42,000 UK premises used Starlink services, up from 13,000 in 2022. The EU shows similar growth patterns, with Germany, France, and the Nordics leading uptake due to large rural areas and digitalization policies targeting underserved regions.

Businesses using satellite broadband span various sectors. In agriculture, farmers leverage satellite internet to run smart farming tools and manage livestock via IoT sensors. Construction firms use it to connect temporary sites. Emergency services and aid organizations rely on satellite connections for dependable communications during crises.

For years, satellite broadband has filled coverage gaps in rural and hard-to-reach areas. But today, its utility is expanding well beyond remote villages. “One of the key commercial use cases for satellite broadband – and where it provides great opportunity – is in the realm of disaster recovery,” says Dave Davis MBE, senior technical director at ST Engineering iDirect. “Even if the fibre or terrestrial connection is lost, satellite broadband would allow emergency services or search and rescue teams to continue their operation.”

Industries such as maritime, aviation, construction, energy, and defense increasingly see satellite broadband as not just a backup but a necessity. Mike Hicks, principal solutions analyst at Cisco ThousandEyes, tells ITPro: "Industries with mobile assets – such as shipping, airlines, and logistics – can greatly benefit. Additionally, energy and utilities sectors may find significant advantages, particularly in remote or offshore locations."

These use cases reflect the satellite's unique strength: its ability to deliver resilient connectivity wherever needed, without the delays or costs of building terrestrial infrastructure. Hicks emphasizes that satellite networks are becoming an essential layer in multi-path connectivity strategies. "It can serve as a hybrid or backup option, particularly within a multi-path, resilient connectivity strategy," he explains.

Yet not all scenarios justify the cost. While Starlink has slashed consumer costs over time, enterprise-grade satellite internet is still relatively expensive compared to fibre or even 5G. "Satellite communications will be more expensive, as far as the monthly rates and things like that go," Davis says. "But then on the other side, you've got an increasing cost per bit, and the hidden costs of cloud computing to factor in."

Connectivity strategies are going hybrid

For most enterprises, the future of connectivity isn’t about choosing between fibre and satellite—it’s about combining them. Satellite broadband is emerging as a critical piece in the hybrid connectivity puzzle, particularly for ensuring resilience. “Many enterprises are adopting hybrid connectivity models that combine satellite with terrestrial networks to enhance reliability and performance,” Dumbleton explains. “This integration allows businesses to maintain connectivity even when one connectivity method experiences issues.”

This concept isn’t new to military and government circles. The term “PACE”–Primary, Alternate, Contingency, Emergency–is a well-known strategy for maintaining communications. Davis says the same logic is now entering commercial conversations. “I think in some cases, it would be primary, particularly rural areas. In other cases, it’ll be a fallback or it could even be a hybrid between the two,” he tells ITPro.

However, there are still hurdles to mainstream enterprise adoption. Service Level Agreements (SLAs), for example, are a sore point. “Particularly with the vertically integrated solutions, there aren’t always any SLAs in place and it’s ‘best effort,’” Davis says. Hicks, however, is optimistic: “LEO networks are changing this perception. Providers are starting to offer more detailed SLAs that include guarantees for uptime, jitter, and even latency.”

Environmental and inclusion goals could drive adoption

Beyond economics, there’s a growing recognition that satellite broadband aligns well with ESG (environmental, social and governance) and digital inclusion goals—especially in Europe, where regulatory alignment and sustainability are strategic priorities. Satellite networks require far less physical infrastructure than fibre or 5G rollouts, making them attractive in regions with rugged terrain or sparse populations.

From a social perspective, the role of satellite in bridging digital divides cannot be overstated. “LEO satellites offer a powerful opportunity to bridge connectivity gaps in underserved and remote regions,” Dumbleton says. “They also play a critical role in supporting emergency services and ensuring continuity of communication during natural disasters.”

However, Dumbleton also cautions about the environmental trade-offs. “Spectrum is a finite resource, and the proliferation of satellites raises the risk of long-term space debris,” he explains. These concerns make coordinated governance and international collaboration essential. Fortunately, the UK and EU regulatory frameworks are largely supportive. “The UK and Europe pretty much follow the approach that if you need it, you will get access to it,” Davis says.

As ESG metrics become central to procurement and investment decisions, satellite broadband’s low-impact deployment could make it an appealing choice—not just for remote areas but also for sustainability-conscious corporations.

Is the future of business connectivity in orbit?

Satellite broadband is no longer a fringe technology. For businesses in the UK and EU operating beyond the reach of fibre, or requiring ultra-resilient connectivity, LEO satellite services offer a viable, though expensive, solution. With Starlink already in play, OneWeb is growing its partnerships, and Kuiper is preparing for launch, so competitive dynamics are heating up.

But one thing is clear: satellite broadband is no longer the outlier. It is becoming a key enabler of digital transformation across sectors and geographies, and it’s well on its way to establishing its place in the enterprise connectivity mix.

Ultimately, it’s not about whether satellite broadband can replace fibre—it’s about where it can extend or enhance what businesses can do, regardless of geography.

Cato Networks has announced a refresh of its global partner program in a move the SASE specialist said will help partners capitalize on market growth.

Available now, the Cato Networks Channel First Partner Program introduces new specialization tracks and program tiers, as well as zero upfront financial commitment, while Cato’s Managed SASE (MSASE) Partner Platform - introduced in June 2024 - will continue to operate as a core framework of the new initiative.

The firm said the program has been built on three core principles to differentiate it from legacy models: transparency, predictable revenue with scalable growth, and maximized gross profit.

Instead of being restricted by legacy frameworks, partners can now benefit from clear pricing, built-in sales tools, and a reliable profitability framework to grow revenue and boost margins over time.

“Our partners asked, and we listened,” said Karl Soderlund, Cato Networks’ global channel chief. “When designing a best-in-class partner program, we landed on four key tenants: predictable profitability, flexibility, simplicity, and differentiation. Put simply, our program reflects how partners want to work today.

“We’re making it radically easier for partners to scale their SASE business with a platform that’s proven to win in the market.”

The Cato Networks Channel First Partner Program

Where legacy programs often feature rigid structures, Cato said its program is more accepting of how partners go-to-market, offering agility, scale, and the opportunity for repeated success.

There are new specialization tracks for each partner type - including VARs, MSPs, service providers, distributors, and technology service distributors - while a new tiered framework aims to deliver greater engagement choice.

Partners can leverage benefits and support across Starter and Advanced tiers depending on their level of engagement and performance.

With zero upfront financial commitment, partners also have the opportunity to sell Cato with no upfront capital expenditure or hidden costs, while benefiting from immediate access to onboarding, certifications, and tools.

Those that sign up to the program will have access to the Cato MSASE Partner Platform, which features a host of certifications, co-branding opportunities, multi-tenancy, and upselling tools with no restrictions on partner type.

Additionally, partners can earn a “Powered by Cato” designation when adopting the Cato SASE platform, while the firm said it will also continue to offer its existing Cato Distinguished Support Provider (CDSP) accreditation for those delivering high-quality support.

“When partners get our SASE platform in the hands of a customer, seven out of ten times, they win,” Soderlund added. “That’s the power of a true SASE platform in the channel.”

MORE FROM ITPRO

• SonicWall CEO hails ‘pivotal moment’ as firm unveils new MSP cyber solutions
• Westcon-Comstor unveils new managed SOC solution for Cisco partners
• Databricks continues EMEA growth drive with double leadership appointment

Businesses are spoilt for choice with network monitoring software, but few can match the track record of Progress Software's WhatsUp Gold (WUG) as it's been a main player in this market for over 30 years. Well-suited to SMBs and enterprises alike, it has consistently offered an impressive range of monitoring tools, all easily accessible from an informative central web console.

WUG gets the benefit of regular updates and new features with version 2024.0.2 on review offering enhanced device template rules, making it easier to identify and define network devices found during the discovery process. Service monitoring and search facilities have been improved while the WUG Live feature presents a more informative status bar and options to customise what events are displayed.

A key new feature is the NTA+ (network traffic analysis plus) option, which takes flow data from Progress Software's FlowMon Collector appliances. Previously, these data feeds were viewed from the standard NTA analysis dashboards, but Progress Software has repositioned this feature with WUG now providing a separate NTA+ menu option specifically for FlowMon data.

Progress WhatsUp Gold Enterprise Plus 2024 review: Licence to monitor

Progress Software is commendably transparent about pricing and offers WUG in a range of flexible plans. It employs device licences, which are easier to manage than sensor-based licences as they allow you to monitor everything on a device regardless of the number of components.

A Business subscription starts at a shade over £900 per year for 50 devices and delivers all key features, including network discovery, topology mapping, alerting and reporting plus cloud, wireless network and storage monitoring. An Enterprise subscription enables log management plus virtualisation host and application monitoring, while Enterprise Plus starts at around £2,600 per year and adds switch, router, and firewall configuration management along with NTA+.

Perpetual licences are also available with Premium costing around £3,600 for 50 devices and providing all standard monitoring services, while Total Plus activates all WUG add-ons except NTA+, which is only available as a subscription. Confusingly, Total Plus prices are based on points where one device costs a point while monitored applications, log management, and NTA flow sources each cost ten points.

Progress WhatsUp Gold Enterprise Plus 2024 review: Deploy and discover

Installation on a Windows Server 2022 host took 30 minutes, and after adding all required device credentials, our first wizard-assisted network scan only took 10 minutes. The discovery process is very accurate as WUG correctly identified all our Windows systems, switches, VMware and Hyper-V virtualization hosts, firewalls, NAS appliances, and printers.

The WUG web console is easy to navigate as all features are grouped under five main menus in an upper ribbon bar, with the Favourites menu in previous versions replaced with a new one for the NTA+ module. WUG gets down to business straight away by automatically applying a base set of active and passive monitors and thresholds to each discovered device.

Color-coded icons are assigned to each monitored element so you can see easily if they are up or down. The home dashboard can be customised to show just about anything you want, while the NOC (network operations centre) feature presents support staff with a heads-up display that combines slide decks of multiple dashboards with a view duration.

Progress WhatsUp Gold Enterprise Plus 2024 review: Action stations

WUG provides plenty of tools for creating alert policies, linking them with device state changes, and assigning actions. You can choose from 24 actions, including sending SMS and email, restarting a service, running a PowerShell script, and linking up with services such as Microsoft Teams, ServiceNow, and Slack.

The Virtualization Monitoring add-on supports VMware and Hyper-V hosts and will discover them from standard scans. From the Analyze tab, we could select a host and view information about running VMs and their vCPU, memory, and virtual interface utilisation.

The application monitor add-on runs discoveries on devices for apps selected from the WUG library and adds them to its analysis dashboard for at a glance status views. WUG's Configuration Management uses scripts to automatically backup the startup and running configs of devices such as switches and routers, and when multiple files are selected in the configuration archive page, it compares them and highlights any differences.

Progress WhatsUp Gold Enterprise Plus 2024 review: NTA+ features

WUG Enterprise Plus includes a licence for the NTA+ Collector, which can be hosted on VMware, Hyper-V, and KVM hypervisors or in the cloud on AWS, Azure, and Google Cloud. This is a scaled-down version of FlowMon Collector, which enables the FlowMon Monitoring Center (FMC) component and is used for network traffic monitoring, managing NetFlow/IPFIX flow sources, and viewing traffic graphs. Three NTA+ Collector versions are available with the base licence activating a 500 VA model, which scales up in step with the monitored device tier you choose to purchase.

We run the full-fat version of FlowMon in the lab, which was automatically identified by WUG and required authentication using OAuth 2.0, Rest API credentials, and a token URL – all clearly documented. From WUG's NTA+ window, we selected our FlowMon appliance and created dashboards by choosing the number of columns and adding reports from the right-hand menu.

There are plenty of reports available, including events, flows, packets, and response times, and you can add custom reports by creating FMC profiles, assigning channels, and setting them as static or continuous for dynamic views. These appear as perspectives in NTA+, and we created FMC profiles for areas of interest such as Microsoft 365, email, web browser, VoIP, and user activity, and selected a display time period – if you want security event feeds, you'll need to purchase an extra FlowMon anomaly detection licence.

Progress WhatsUp Gold Enterprise Plus 2024 review: Is it worth it?

Oh, yes. WhatsUp Gold delivers an incredible range of network monitoring tools for the price and amalgamates them into a single management console, offering deep insights into all devices, services, and applications.

The Enterprise Plus version on review is particularly good value as it enables all the add-ons Progress Software has to offer. More clarification on the various NTA+ Collector licences would be helpful, but with FlowMon behind the scenes, there's no denying this is capable of providing a wealth of information that will be invaluable for troubleshooting network issues.

HPE has announced a sweeping expansion of its Aruba Networking Central platform, offering users a raft of new features focused on driving security and data sovereignty.

As part of the move, users of the network management solution will be granted access to a virtual private cloud (VPC) environment aimed at shoring up data security and meeting regulatory requirements, the company said.

This will be complemented by an on-premises option capable of operating while disconnected from the cloud.

Phil Mottram, EVP and general manager for HPE Aruba Networking, said the expansion comes in direct response to evolving enterprise needs with regard to data sovereignty.

Enterprises operating on both sides of the Atlantic, particularly in the European Union (EU), face strict rules governing the use, storage, and sharing of customer data.

“Organizations are increasingly prioritizing data sovereignty, requiring regional and local presence for mission-critical IT solutions,” he said.

“With these innovations, HPE now uniquely addresses the most pressing enterprise challenges for corporate, nonprofit, and government entities with unprecedented network management deployment flexibility.”

HPE Aruba Networking Central changes

As part of the expansion, HPE Aruba Networking Central will gain four new deployment options. This means the solution can now be deployed in a cloud-delivered SaaS, VPC, on-prem, or NaaS format.

Similarly, integration of FIPS 140-2 certified hardware aims to help enterprises meet government security requirements and adhere to regulations including GDPR.

Described as a “comprehensive networking management solution”, HPE said this will boost efficiency for use-cases such as AI data capture, training, and inferencing - all of which typically require greater levels of control through air-gapped on-prem and cloud-based VPC options.

Underpinning this is the company’s GreenLake platform, which will provide enterprise users with local cloud options across a range of locations.

The firm is also keen to highlight gains made with regard to global points of presence (PoP), with dedicated instances now hosted in the US, Canada, EU, Middle East, Africa, Asia Pacific, and China.

AIOps tools look to drive network productivity

Elsewhere, other new features unveiled by HPE include the launch of automated network AIOps capabilities.

This will enable enterprises to continuously monitor critical wires and wireless network operations, optimize processes, and detect performance issues.

“A fabric of AI assistants acts as network architects, monitoring and gathering data, providing diagnostics and recommendations to extend existing capacity and performance, close security gaps, and identify configuration errors before they impact network operations,” the company said in an announcement.

“Combined with our continued innovation across AI, security and connectivity, HPE Aruba Networking Central continues to offer the most powerful and versatile network management application on the market, helping organizations meet their security, privacy and control requirements,” Mottram said.

MORE FROM ITPRO

• HPE unveils Mod Pod AI ‘data center-in-a-box’ at Nvidia GTC
• What HPE's results say about the direction of enterprise AI
• HPE’s ‘one-click AI solution’ for private cloud cuts project times from months to a ‘single moment’

Research shows the vast majority of Wi-Fi networks are vulnerable to a popular type of denial-of-service (DoS) attack that is frequently deployed in larger cyber intrusion efforts.

A new report from Nozomi Networks that analysed telemetry from hundreds of OT and IoT environments found 94% of Wi-Fi networks lacked the proper protections against deauthentication attacks.

Deauthentication attacks are a form of DoS attack that targets openings in network protocols to force devices off the network and disrupt operations.

The report noted that these attacks are often leveraged in the opening stages of larger, more devastating attacks, softening up the organization’s defenses for future attempts to infiltrate their network.

“They leverage a built-in feature in the Wi-Fi protocol, specifically in the management frames used for communication between devices and access points. By transmitting fake deauthentication frames, attackers can force devices to disconnect from the network,” the report explained.

“This can escalate into more severe disruptions, such as data interception and unauthorized access, especially when combined with additional malicious actions.”

Nozomi added that the crux of the problem lies in that only 6% of the observed wireless networks featured management frame protection (MFP), which it describes as a crucial security feature that prevents attackers from spoofing management frames.

This means that virtually all networks, including those that underpin critical national infrastructure (CNI), are vulnerable to malicious attacks, the report warned.

“The vast majority of wireless networks, including those in mission-critical environments, remain highly exposed to these kinds of attacks. In healthcare, for example, vulnerabilities in wireless networks could lead to unauthorized access to patient data or interference with critical systems,” Nozomi said.

“Similarly, in industrial environments, these attacks could disrupt automated processes, halt production lines or create safety hazards for workers.”

Wireless security is a critical priority

The report warned that in light of recent attacks from state-linked threat groups targeting networks at CNI organizations, wireless network security has emerged as a critical factor in bolstering resilience.

In recent years, groups like Volt Typhoon and Salt Typhoon have been responsible for compromising network infrastructure at organizations across a range of critical sectors in the US and further afield.

The groups are linked to attacks on the US telecoms industry, where the attackers are said to have recorded conversations of very senior political figures, as well as established persistence on the networks of organizations that make up the US electricity grid.

The report highlighted a number of the other common threats used to target wireless networks. For example, attackers can deploy rogue access points to impersonate legitimate networks and trick devices into connecting to them and exposing sensitive information.

Jamming attacks are another popular method where malicious actors overwhelm wireless networks in order to force downtime and operational disruptions.

The report also warned eavesdropping attacks, where communications on unencrypted wireless protocols are intercepted, allow threat actors to steal credentials, read sensitive data, or monitor operations.

Nozomi emphasized that the modern threat landscape requires a shift from static to dynamic security measures, which includes implementing a risk reduction strategy, prioritizing anomaly detection and response, strengthening endpoint security, and applying network segmentation to limit the reach of attacks from botnets.

Organizations should strengthen wireless network security with regular audits and this should include prioritizing mitigating common threats, such as deauthentication attacks.

MORE FROM ITPRO

• UK cyber experts on red alert after Salt Typhoon attacks on US telcos
• FCC orders telcos to sharpen up security after Salt Typhoon chaos
• Wireless networks are a prime target for hackers – here’s how to combat rising threats

Internet service providers (ISPs) have evolved significantly in their role as key enablers of business success, growing alongside the sky-rocketing demand from organizations to take full advantage of the power of the internet. 

Despite their relatively humble beginnings, ISPs have grown to become behemoths in the world of technology, on which much of the industry relies to provide their services.

Now, Artificial Intelligence (AI) and the Internet of Things (IoT) are both expected to place unprecedented demand on ISPs’ infrastructure. As these new technological advancements gain momentum, ISPs must now continue to adapt and evolve alongside the market. 

To look forward, though, it’s important to take a little look back…

Step back in time

The first commercial ISPs emerged in the late 1980s and early 1990s, with the rise of dial-up as the first form of internet access that saw widespread uptake among ordinary people. Those households around the world that already had computers were purchasing modems and using software provided by the ISP to dial into the connection via their phone lines.

Although impressive at the time, it was a fairly rudimentary solution with very limited functionality. For example, the household could not use their phone line to make calls while using the internet, which was not very practical.

As such, the expectations placed on the ISP were relatively limited, offering basic connectivity with no firm promises on the speed of your connection or any additional services. 

Speeds topped out at 56kb/s so the most strenuous tasks these connections could handle were stripped back to email and file transfer (which itself was fairly limited due to the speed). Downloading any type of media was a painful process, but the limits of the dial-up connection ISPs could do little to improve their customers’ experience.

Security or any advanced functionality was not a massive priority for the ISPs operating in this nascent market, who were mainly catering for enthusiasts excited to get on the web for the first time, but this would soon change.

ISPs’ current role in business connectivity

The arrival of broadband in the 2000s expanded the range of connections to cable and fibre optic, which vastly increased the potential speeds available to end users. This fired the starting gun on expanding what the internet could do for consumers and businesses alike. 

High-speed, dedicated business networks meant businesses could take advantage of exponentially faster, more reliable internet that has now become vital for most organizations’ operations.

The emergence of cloud computing, facilitated by this leap forward in connectivity technology, ushered in the era of the as-a-service model, where businesses can make use of infrastructure, platforms, and software through subscription models without having to stump up exorbitant up front costs or pay to maintain them.

This model has proliferated rapidly across sectors and verticals, leaving the majority of businesses reliant on their ability to connect to their cloud providers, as well as the ever-increasing number of other service providers this revolution has created. 

As such, the ISPs have entered the realm of critical infrastructure underpinning the operations of many essential services, meaning the reliability of the connection has become the highest priority.

Any disruption in connectivity could impact their productivity, customer service, and bottom line, so businesses expect a fast and reliable internet service with minimal downtime for their day-to-day operations. 

Part of this puzzle involves having access to fast and effective customer service when things do go wrong, which means ISPs are often expected to deliver round-the-clock dedicated business support teams with quick response times. 

A reliable connection encompasses the security of the devices using it and the data passing through it. Businesses handle various forms of sensitive data and are frequently the target of increasingly sophisticated cyber attacks. 

Accordingly, they expect ISPs to offer a number of robust security features such as firewalls, threat monitoring,  spam filtering, DDoS defenses, and intrusion detection features.

Faster, more secure connections for the intelligent era

One area in which businesses are continually placing higher expectations on their ISP is increasing bandwidth as they scale. 

As enterprises grow and become more digitally proficient, they will look to adopt the latest and greatest tech for their organizations, and in this era that will almost inevitably place new burdens on their ISP.

AI, for example, has received huge interest from businesses of all sizes looking to implement it in all its forms across their operations, but having a strong connection is often vital, particularly if the business in question is one of the many that won’t have an on-premise solution.

ISPs will be expected to not only provide the low latency, ultra-high bandwidth connectivity to help businesses leverage their AI deployments, but also use AI to improve their service. 

This includes using AI to optimize their network management, enhance customer service with LLM-powered chatbots, and provide expanded security capabilities with enhanced threat detection.

In a similar vein, as businesses embrace other technology trends such as IoT, edge computing, and hybrid working models, ISPs will play an increasingly critical role in providing the low latency, ultra-high bandwidth connectivity to support their hyper-connected infrastructure.

To cater to this new demand for performance and scalability, ISPs will need to invest in high-capacity networks that can handle this increased throughput with minimal latency.

As cloud integration becomes an integral part of almost every organization, ISPs will have to work to build out streamlined integrations with all the major cloud providers, as well as cater for increasingly common multi-cloud and hybrid cloud environments.

Evolution to support changing demand

To support this, ISPs will need to ensure they have robust data center infrastructure to make sure they can offer reliable and fast cloud services that cater for the large-scale data storage and retrieval tasks their customers require.

Another consequence of businesses accelerating their digital growth for the intelligent era is that it introduces new openings for malicious actors to compromise their IT environments and cause damage by exfiltrating data or simply causing a business shutdown.

This means ISPs will need to prioritize cybersecurity infrastructure including the aforementioned AI-based threat detection but also advanced network security such as zero-trust models, identity and access management, and enhanced network monitoring.

Finally, Regulations like DORA, GDPR, and NIS2 have levied new pressure on businesses to take control of their operational resilience and data protection. This means ISPs must also be aware of and comply with global and local regulations, especially when handling business data. Looking forward 

Overall, as the industry continues to be shaped by an era that is more connected than ever before, ISPs have become increasingly crucial pieces of critical national infrastructure. Their responsibility extends far beyond simply connecting users to the internet, but managing this connection and providing an array of additional services businesses need to get the most out of it.

Bandwidth and speed have never been more important as new technologies like cloud computing, IoT, and AI, place unprecedented demands for high data throughput and low latency, while guaranteeing the security and privacy of that traffic is also becoming a growing priority for businesses.

By focusing on the infrastructure considerations laid out above and aligning with the business needs as they evolve to match the new technological landscape, ISPs will continue to be essential business enablers facilitating the digital transformation and underpinning success in an increasingly connected world.

AI can help operators accelerate broadband deployments, in particular if it's used to design networks, according to new research.

A report by the World Broadband Association (WBBA) argued that the technology could help in the pre-deployment phases of broadband networks by being put to work during network planning and expansion.

The findings of the report were presented at the Broadband Development Congress (BDC) in Barcelona this week.

So far, telecom companies are focused on AI for customer service and network troubleshooting and optimization — all useful applications but by bringing AI into systems earlier, there could be further benefits.

By using AI to help with network development, operators can make sure they roll out infrastructure to the right places at the right time and with the right technologies, the report noted. Similarly, this could help ensure a return on investment (ROI).

"The pre-deployment phases in broadband are periods of decision-making and network planning before infrastructure is rolled out or upgraded," said Martin Creaner, Director General of the WBBA.

"It’s an ideal environment for AI to make an impact, and if its models are supported by the right data and algorithms, then it can be used to help operators predict the return on their investments much more accurately."

Using AI to support broadband deployments

The white paper suggested, for example, that an AI model can identify the most suitable areas for a new broadband deployment, when to make upgrades from copper to fiber, or how to improve penetration in rural areas by considering existing networks, customer data, and geography, as well as costs.

"All of these decisions should be informed by many interdependent factors and moving pieces related to deployment costs, demand dynamics, wholesale versus retail trade-offs, and regulatory constraints," the report noted.

"AI can play a key role in generating and analyzing a huge quantity of high-quality data for some (or all) of these factors to produce outputs for one (or more) of these decisions while optimizing the ultimate objective variable (which, in this case, is assumed to be ROI in broadband investment)."

That information can help operators decide when and where to deploy, which customers to target, and how to price services, the white paper noted.

However, the WBBA warned that this approach requires the right dataset and that AI has limitations. As such, operators should understand that it works best with clear constraints and is used alongside existing analysis for the best results.

"AI is not a magic wand that can solve the entire ROI optimization problem, but it can add significant accuracy to parts of the equation," the paper noted.

AI needs broadband too

The WBBA also noted that more efficient broadband was necessary to power the shift to AI and next-generation internet services.

"There is no single use case that will drive demand for gigabit broadband access," the WBBA said in a separate report about Gigabit broadband.

"It will be driven by a combination of three core trends: a rapid increase in connected devices, the development of advanced video-enabled AI applications, and the shift of computing power to the cloud. This combination will drive the need for high-speed, low-latency, highly reliable, and consistent networks."

As part of its Gigacity Index 2025, the WBBA ranked the broadband sophistication of cities around the world, revealing Singapore, Shanghai, Zurich, Hong Kong, and Dubai as the leaders.

MORE FROM ITPRO

• UK rural businesses set for broadband boosts
• Over half a million UK businesses struggle with insufficient bandwidth
• Poor broadband connectivity is costing London SMBs billions

Security experts have issued a warning over surging threats faced by wireless networks, with many left woefully unprotected .

Analysis of more than 500,000 'honeypot' wireless networks by Nozomi Networks Labs found only 6% are adequately protected against wireless deauthentication attacks.

The rest are vulnerable to unauthorized access to data or interference with critical systems, the study noted. For example, in industrial environments attacks could disrupt automated processes, halt production lines, or create safety hazards for workers.

"Cyber attacks on the world’s critical infrastructure are on the rise," said Chris Grove, director of cybersecurity strategy at Nozomi Networks.

"The systems we design and defend must not only withstand a barrage of threats in today’s multipolar world but also balance the need to operate safely at scale, where human lives are at stake."

In the second half of last year, nearly half of observed cyber threat alerts took place during the ‘impact’ phase of the cyber kill chain. This was true across various industries, but particularly in manufacturing, transportation, energy, utilities, and water.

Next were Command and Control (C&C) techniques, which accounted for a quarter of all observed alerts.

Meanwhile, of the 619 new vulnerabilities published in the second half of 2024, 71% are classified as critical. Another 20 have high Exploit Prediction Scoring System (EPSS) scores, indicating a high likelihood of future exploitation.

Indeed, four vulnerabilities have already been observed being actively exploited in the wild.

Of all ICS security advisories released by CISA over the past six months, critical manufacturing topped the list, accounting for three quarters of all Common Vulnerabilities and Exposures (CVEs) reported.

Next were manufacturing, energy, communications, transportation, and commercial facilities.

The energy industry - a lucrative sector for threat actors - is becoming a growing target, with 174 attacks taking place since 2024; more than half were focused on credential access and collection.

Across the board, common threats include deauthentication attacks, rogue access points, eavesdropping, jamming attacks, signal interception and espionage, signal jamming and network access attempts.

How to counter network threats

According to Nozomi Networks, organizations should implement a risk reduction strategy and audit performance regularly, prioritizing anomaly detection and response.

"With nation-state threat actors relying on living-off-the-land tactics and malware inflicting multimillion-dollar losses on victim companies, keeping threat intelligence databases up to date is critical to safeguard against known threats," the researchers warned.

"However, signature-based detection alone may not suffice for emerging or unknown threats. Many reconnaissance activities are more effectively uncovered through anomaly detection rather than purely signature-based approaches."

It's also important to take note of regional and industry-specific threat intelligence and prioritize vulnerabilities with high-risk scores, the study noted.

Organizations should also take asset criticality, compensating controls, device type, safety implications, exposure and other contextual factors into account.

Similarly, defenses against botnet attacks should be improved by strengthening endpoint security and applying network segmentation.

"By understanding these evolving threats and leveraging actionable insights, we can defend our critical infrastructure systems to ensure resilience, safety and operational continuity in an increasingly uncertain world," said Grove.

MORE FROM ITPRO

• Global cyber attacks surged 44% last year
• Manufacturing firms grappling with growing OT threats
• Five real-world cyber attacks and how to stop them

Networking automation provider Extreme Networks has announced a new program for MSPs, which includes access to its Extreme Platform ONE to address modern cost and billing challenges.

Available from March 31, the firm said the initiative features the industry’s first and only consumption-based billing model and has been designed to equip MSPs with increased flexibility, simplicity, and cost savings.

The new framework has been built on AI-driven, multi-tenant architecture and now features consumption-based billing alongside poolable licensing.

Extreme’s MSP program also enables partners to offer networking, security, and AI solutions that help scale operations and create differentiation in a crowded market.

In an announcement, Extreme Networks’ chief product and technology officer, Nabil Bukhari, said that, by removing the barrier of hefty upfront costs, more MSPs will be able to adopt the firm’s model - ultimately speeding up time to market and revenue.

“Our consumption license grants full control over distribution, activation, and management, while Platform ONE boosts productivity by unifying everything into a single collaborative workspace,” he explained.

“This streamlined approach empowers MSPs to drive efficiency and growth with greater speed and simplicity.”

Extreme Networks eyes automation gains, consumption-based billing

MSPs will be able to leverage Platform ONE to manage licenses, performance, and service interruptions from within a single, centralized workspace that offers a comprehensive view of each client’s network health, performance, and security status.

The platform also serves up automation through conversational, interactive, and autonomous AI agents to drive productivity and condense the time it takes to complete complex tasks “from hours to minutes,” Extreme said.

Extreme’s Platform ONE’s consumption-based billing model aims to ensure cost predictability, with poolable licensing allowing MSPs to dynamically allocate and reassign licenses across devices, locations, and customers for increased flexibility.

Cara Parfitt, vice president of technology alliances at Logically, described Extreme’s MSP program and consumption-based model as a “game changer” for the company.

“The consumption-based billing model has provided us with more predictable costs, allowing us to better manage our finances and reduce the risk of our IT investments,” she commented.

“With Extreme Platform ONE, we can efficiently manage all our clients under one platform, which will significantly reduce the time and IT staff required to maintain operations.

“This will enable us to scale our operations, diversify our service offerings, and ultimately deliver better value to our customers."

MORE FROM ITPRO

• CyberArk eyes identity security gains with $175 million acquisition
• Thales and Imperva unveil aligned global partner programs for 2025
• Red Hat launches Build module as part of partner program refresh

Last year, Cisco Live 2024 was my first proper introduction to the company in all its glory and if I’m honest I was somewhat overwhelmed by its scale and breadth.

Fortunately for me, the overriding theme in Amsterdam that year was tightly focused on unifying Cisco’s byzantine product ecosystem under one integrated platform. In 2024, the firm’s foremost storyteller Jeetu Patel, who was EVP and GM of security & collaboration at the time and now is Cisco’s chief product officer, set out his strategy for how Cisco was going to drive value for customers through its platform approach.

I found this to be a compelling vision. Businesses can derive significant value from integrated platforms that help them manage the complexity of their IT estate, as they rush to add various technologies to keep their services efficient and secure while contending with skills shortages that limit their ability to do so.

Fast forward to Cisco Live 2025 and this message was still there. But while simplifying complexity through integration remained a key undercurrent in many of Cisco’s announcements, the focus on the platform felt more implicit this year rather than the central theme of the event. This year’s announcements showed that while the unified approach is important to Cisco, it’s the quality of the networking and security products being unified that’s really the key to its success.

Throughout the event, Patel stressed the importance of the company’s networking and security announcements in the context of Cisco’s wider ecosystem. Take the new N9300 Series Smart Switches, for example, built using Cisco’s Silicon One architecture and now featuring AMD’s Pensando data processing units (DPUs).

These new switches aren’t the first to be equipped with DPUs but Cisco says their unique value comes with the combination of these switches and its existing Hypershield security architecture, something competitors can’t offer.

By combining of these two offerings promises to bring advanced security inspection and segmentation services across a distributed network, giving end-users the ability to embed security at the micro service level.

The convergence of networking and security is something Cisco has always had a strong focus on and this marks a significant milestone towards that end.

Its first forays into networking security were a natural progression from pure networking products] and since then the firm has relentlessly pursued becoming a fully fledged security company with a series of acquisitions. Integrating these security services into the networking side of the business has been a significant focus for Cisco over recent years, and now it looks like this is coming to fruition.

On stage Oliver Tuszik, EMEA president at Cisco, stressed the old adage that today every company is a tech company. Giving it a 2020s twist, he added that every company is now also becoming an AI company – making digital resilience the other side of the coin from business resilience.

Considering the fact that the network is the backbone of any IT estate, consolidating security and networking has always been an easy decision for Cisco, and now it can look to reap the benefits.

Cisco’s bread and butter is still the network – especially for AI

With the complications of AI advancements such as DeepSeek and Operator (OpenAI’s AI agent) will put new burdens on IT teams. Patel predicted that businesses could be using hundreds or thousands of models, and multiple AI agents in multiple different data centers, which will place unprecedented strains on networking infrastructure.

Across the course of the event, Cisco didn’t shy away from showing off a stream of products they were bringing out to address this fact. A new AI server, the UCS C854A M8 fills out a previously missing part of Cisco’s AI data center portfolio with a flexible solution dedicated for training large models.

The company’s Agile Service Networking architecture, Cisco 8000 Series routers and pint-sized Coherent Pluggable Optics, also serve the same aim of helping customers modernize their networks ready for the oncoming wave of AI workloads they envisage running in the near future.

Speaking to ITPro, Bola Rotibi, chief of Enterprise Research at CCS Insight, agreed that this year the explicit references to the platform took a back seat as the firm leant back on its networking bona fides, but she said this was a shrewd move in many ways and highlighted where Cisco really has an edge on the competition.

“I think they’ve made some very smart acquisitions but at the same time I think where they have not lost their edge is that they are the network company and that everything they think about has to be framed in how we help people with all the other pieces," she explained.

“So the story about the network driving AI is a really smart move because it allows them to talk about various technologies from their strength and their focus on the network, I think that’s a really smart move.”

This isn’t to say the platform was completely sidelined, however. Patel cited the platform during the opening as a structural advantage Cisco has that it can use to capitalize on key technological shifts this year, namely DeepSeek’s seismic impact on the AI industry as well as the ongoing fervor around AI agents in the enterprise.

Patel described it as Cisco’s ‘platform advantage’ that ensures that as customers invest in new infrastructure according to these two macro trends they see a compounding value of the platform.

“Every single capability and product that we build should reduce the marginal cost of ingestion of every subsequent product that’s actually going to be ingested by you.”

“We want to make sure that the cost of adopting new technology from Cisco goes down precipitously and it compounds value in the products that you might already have.”

Both Tuszik and Patel expressed that they were “bullish” on the firm's position in the market moving forward, but with financial disclosures to come in the days following the conference they couldn’t elaborate.

It looks like they were right to be confident, however, as the firm raised its annual outlook citing strong demand for AI networking infrastructure with its stock jumping to a 24-year high.

My only concern last year around Cisco’s platform ‘story’ was that there was no mention of how other vendors would fit into this picture, and the company did little to reassure me this year either. The IT ecosystem is defined by heterogeneity and no matter how big, Cisco will need to demonstrate how its platform connects to business solutions from other vendors, and how it will help customers manage this complexity.

MORE FROM ITPRO

• Cisco wants to capitalize on the ‘DeepSeek effect’
• Cisco is jailbreaking AI models so you don’t have to worry about it
• Cisco dispels Kraken data breach claims, insists stolen data came from old attack
• A CIO's trifecta in edge networking: AI, 6G, and security

HPE has hit out at the US Department of Justice (DOJ) over its attempt to block the acquisition of Juniper Networks, suggesting the move is “divorced from reality” and will benefit competitors in the space.

HPE bid $14 billion for Juniper Networks in January 2024, with the deal seen as a play by the networking giant to take advantage of the AI boom. With an eye to market consolidation, European and UK regulators examined the deal, but the acquisition was approved in August.

Despite gaining clearance in European markets, the DOJ lodged a legal challenge to block the acquisition, specifically highlighting the WLAN supply market and arguing that cutting competition will harm innovation.

The legal challenge claimed that by cutting three major players to two - these being HPE, Juniper, and Cisco - the deal will limit options for customers and ultimately lead to higher prices.

At the time, HPE and Juniper released statements disputing the DoJ claims, suggesting the WLAN market was more robust than the lawsuit argued.

Now, in a document filed in response to the DoJ lawsuit, HPE said its tie up with Juniper would give customers a "credible alternative to Cisco", the networking giant that dominates the sector.

Beyond Cisco, HPE points out that Chinese networking giant Huawei is considered a security risk by the US government, so it's in need of other options for infrastructure to resist the use of Chinese technology globally.

The filing says the suit from the DoJ will "will hobble competition with Huawei — which has been repeatedly identified as a national security risk by the U.S. government — and thus damage the US’ stated aim of reducing the use of Chinese technology in critical infrastructure globally."

Why WLAN?

The HPE filing noted that HPE and Juniper are two of ten companies that battle it out in the WLAN space, saying the DOJ description of HPE's dominance in the space is "divorced from reality".

“This is a deal that enhances competition and creates a real challenger to Cisco's dominance,” HPE added.

Indeed, it accuses the original complaint of cherry picking data by describing HPE, Juniper, and Cisco as having 70% share of that market. However, HPE said the DOJ is "conspicuously silent" about the respective market shares, as Cisco holds more than half of the WLAN market and has done so for more than a decade.

HPE added that the regulator is mistaken in its focus on WLAN, explaining that the real motivation behind the acquisition of Juniper is combining its data center routing tools with HPE's storage and compute.

"While WLAN is a component of the overall transaction, it is misleading to suggest that HPE is spending roughly $14 billion to acquire Juniper for the purpose of insulating itself from WLAN competition in the United States, particularly when the WLAN solutions that are the focus of the Complaint comprise only 11% of Juniper’s revenue," the filing noted.

"There are simpler — and significantly cheaper — alternatives for HPE to acquire a single digit market share in the United States in WLAN if that was its primary goal."

In its own defense, HPE's filing quoted a LinkedIn post by industry analyst Shamus McGillicuddy of Enterprise Management Associates.

"The DOJ complaint ignores all other aspects of this deal, including data center switching, routing, firewalls, SD-WAN, network automation. It doesn’t even mention Junos, which execs from both companies call ‘the crown jewel’ of Juniper. The press release calls Juniper a ‘wireless LAN vendor.’ Hilarious."

MORE FROM ITPRO

• Juniper Networks unveils revamped partner program
• “Our cloud works”: Juniper Networks exec fires shots at Cisco’s networking architecture
• HPE bid for Juniper Networks sets it up for battle with Cisco

As businesses take a 'wireless-first' approach to network design, the adoption of Wi-Fi 7 is rapidly picking up pace. All the key manufacturers now offer a wide range of Wi-Fi 7 access points (APs) and Ubiquiti Networks stakes its claim with a family of no less than eight APs.

The three UniFi enterprise models are aimed at very high density environments with price tags to match while the five other models offer businesses a more affordable route to Wi-Fi 7. Along with two outdoor and one wall-mounted version, you have the top-dog indoor U7 Pro Max, and here, we take a closer look at the entry-level U7 Pro.

Ubiquiti Networks UniFi U7 Pro review: Design and specs

The U7 Pro is well-built with a solid polycarbonate cover and metal chassis. The box includes metal mounting plates and a complete toolkit with all the components needed for wall mounting and all types of ceiling – even the mounting template card has a built-in spirit level.

At this low price point there are always hardware compromises and the U7 Pro is no exception. Whereas the U7 Pro Max has a big BE15000 rating and 4x4 MU-MIMO, the U7 Pro sees this drop to BE9300 and 2x2 MU-MIMO for each radio.

The U7 Pro presents a 2.5GbE multi-Gigabit port supporting an 802.3at PoE+ power source which you'll need as the AP doesn't have a socket for an optional power adapter. Note also that the AP doesn't have the dedicated radio in the U7 Pro Max that it uses for real-time spectral analysis.

Early U7 Pro models had an internal blower fan fitted behind a small slot next to the network port but our sample has a later board revision which doesn't appear to have this. We certainly weren't going to try and dismantle the AP but if there was a fan, we didn't hear it during testing and the back of the AP never got hot enough to cause any concern and was significantly cooler than the U7 Pro Max model we also had in the lab.

Ubiquiti Networks UniFi U7 Pro review: Deployment

The U7 Pro can be deployed in standalone mode but you'll be missing out on a wealth of features if you go down this route. Remote management options are incredibly diverse as you can install the free UniFi Network Server for Windows, macOS, and Linux, use Ubiquiti's chargeable cloud-hosted controller, or choose from a range of UniFi Cloud Gateways.

We used the lab's UniFi Dream Machine Pro (UDM Pro) appliance which costs around £330 and teams up wireless network management with gateway security plus video surveillance, motion detection, and recording services. We could view all remote gateways from our Ubiquiti Site Manager cloud portal and selecting the UDM Pro transported us to its own management console.

This is a joy to use as it provides everything you need to know about your wireless networks. It shows detailed topology views, lists of all managed UniFi devices and deep insights into connected clients with a complete traffic breakdown along with icons showing the top detected apps and their individual bandwidth usage.

It only has non-PoE Gigabit LAN ports so we used our Zyxel XS1930-12HP 10GbE PoE++ switch which the UDM Pro recognised and added to its topology view. Deploying the U7 Pro Max was a piece of cake as it appeared in the console as soon as it came online and after adoption, it received a firmware upgrade and then presented all our predefined site SSIDs.

Ubiquiti Networks UniFi U7 Pro review: Easy management

Choosing the Control Plane option on the portal's home page transports you directly to all network-related settings. From here, you can create multiple SSIDs, assign radios, and choose from a wide range of options including the preferred encryption scheme, client isolation, rate limits, and an enhanced IoT feature that restricts the network to the 2.4GHz band if you have device connectivity issues.

From the Radio Manager page, we enabled the Wi-Fi 7 320MHz channels but noted that the MLO (multi-link operations) feature is yet to be supported. We understand that firmware version 7.1 will enable this and you can use the portal's advanced system settings to sign up to Ubiquiti's early access program.

Captive portal features for wireless guests are outstanding as you can assign them to selected SSIDs and customize the landing page with your own logos and colors. Authentication methods are plentiful as you can use Facebook, a global password, and a variety of payment gateways including Paypal, voucher codes, or Radius servers.

Network monitoring services are simply the best with graphs provided for WiFi health plus views of airtime, connected clients, and performance stats for the selected AP. The Insights page reveals plenty of details on detected apps and we love the topology view which shows how each device is connected with the internet traffic monitor presenting animated real-time flow lines between each device so you can easily see who is using the most bandwidth.

Ubiquiti Networks UniFi U7 Pro review: Wi-Fi 7 performance

For performance testing, we called up a Lenovo desktop client running Windows Pro 24H2 and equipped with a TP-Link Archer TBE550E Wi-Fi 7 PCIe adapter. Using Microsoft's NTttcp utility, we recorded raw upstream and downstream rates of 235MB/sec and 197MB/sec between the Wi-Fi 7 client and a Windows server on our 10GbE LAN.

Real-world performance was noticeably lower with close-range large file copies between the client and server returning top average speeds of 185MB/sec. To test range, we moved the AP 10 meters away into an adjoining room and recorded top copy speeds of 173MB/sec. These numbers are all on par with similarly priced Wi-Fi 7 APs we've tested.

Ubiquiti Networks UniFi U7 Pro review: Is it worth it?

With a sub-£150 price tag, the U7 Pro sits well with the competition at the entry-level end of the Wi-Fi 7 AP market. Zyxel and TP-Link, for example, offer their respective NWA130BE and EAP772 APs for around £133 which both returned a very similar performance in our lab tests.

Few can match Ubiquiti for remote management features though. You will need to factor in the cost of a UniFi controller but you'll be rewarded by an absolutely stunning range of cloud management and wireless network monitoring services.

Ubiquiti Networks UniFi U7 Pro specifications

DeepSeek’s shock introduction has had a massive effect on the adoption and development of generative AI models in the business landscape, and Cisco feels it has a significant advantage in capitalizing on this.

Speaking at Cisco Live 2025, its annual EMEA conference in Amsterdam, Jeetu Patel, EVP and chief product officer at Cisco said this is a massive opportunity for the firm from both a network infrastructure and security perspective.

Patel said the arrival of DeepSeek was one of two “seismic developments” that have taken the industry by storm, the other being OpenAI’s foray into the world of agentic AI with its Operator solution.

The DeepSeek effect, according to Patel, is that the cost curve for AI deployments has plummeted in terms of the resources it takes to train a model, and as a result the number of models in use in business is about to skyrocket.

“You won’t just have tens of models, you might have thousands of models and this will actually have massive implications on security and safety,” he predicted.

This broad industry shift will expand every organization’s attack surface, according to Patel, introducing a host of new potential threats and piling added strain on businesses’ IT infrastructure.

Cisco aims to capture new industry demand in both of these areas, firstly in the infrastructure space with a new AI server, the UCS C854A M8.

This is a modular system based on Nvidia’s MGX reference architecture that can accommodate anywhere from two to eight GPUs according to the user’s needs and budget constraints.

The UCS C854A M8 is available in a number of configurations, supporting a range of Nvidia GPus and CPUs from AMD, and will fill a gap in its AI infrastructure portfolio for model training.

With the new UCS Cisco looks to fill out its AI infrastructure, joining its recently launched AI PODs solution, aimed at providing flexible, cost efficient generative AI deployments.

Network and security expertise will be essential in managing safe AI deployments

In a panel session following the opening keynote, Patel said the implication of the joint DeepSeek and Operator effects is that business networks are going to have to deal with a whole different new volume of traffic.

He raised Cisco’s in-house network architecture, Silicon One, as a key area where Cisco is well-positioned to exploit new demand AI workloads will place on businesses.

Another aspect to the industry’s ongoing agentic shift is the fact that ensuring these systems are secure and managed effectively.

In addition to its existing network monitoring and access solutions, Cisco recently released its own generative security solution, Cisco AI Defense, that promises to safeguard both the development and use of generative AI applications in the enterprise.

The system can restrict access to external AI applications, helping reduce risks posed by shadow AI, or monitor and manage how users are engaging with these platforms.

On the other hand, Cisco AI Defense has a vulnerability testing function that can identify if your organization’s internal AI applications have any security weaknesses or privacy concerns.

Speaking to ITPro, DJ Sampath, VP of product for AI software and platform at Cisco, said customers are increasingly looking at using both external AI applications and deploying these systems internally. As such, an end-to-end system that covers both of these spheres is vital.

He argued that as firms look to take advantage of open source models available on platforms like Hugging Face, it’s integral that these organizations are able to put the right security, safety, and privacy guardrails in place before deploying them inside their environment.

Sampath claimed it is Cisco’s network security expertise and its new strengths in securing AI systems that puts it at a significant advantage for the company in light of these recent transformative events.

“The biggest advantage that Cisco is going to have is, from an AI safety and security perspective, we’re able to position ourselves where we’re seeing everything on the network, we’re enforcing everything on the network. We have the fabric of the network to do that,” he explained.

“In addition to enforcing things at scale, we are validating these models at scale. So both of those [capabilities] provide us with a very unique way of being able to provide value to our enterprise customers.”

MORE FROM ITPRO

• Cisco Live EMEA 2025: All the news and updates as they happen
• Stargate project: OpenAI, Oracle pledge support for $500 billion AI infrastructure drive
• Huawei targets ‘intelligent industrial transformation’ with AI-ready infrastructure