Every office comes with its own unique quirks — we all know this, of course. They might include dodgy windows that won’t close properly, meeting rooms with broken webcams, or an unnecessarily loud colleague. But spare a thought for Arturs Banks and Rajeev Bhandari, head of IT and third-line infrastructure analyst, respectively, at Fulham FC, whose workplace environment is highly challenging. 

Managing the IT and cyber security needs of a professional soccer club is challenging for a number of reasons, chief among which is the ‘office’ layout. There aren’t many headquarters in the world, after all, that’s cleaved out its core, forcing staff to take the longest route possible when traveling from A to B.

The geography of Craven Cottage, home of Premier League stars such as Alexander Mitrovic and Andreas Perreira, is vastly different from that of, say, Apple’s Cupertino campus. The layout of Banks and Bhandari’s office — a 25,000-seat stadium in West London — they say, proves a major headache.

The soccer stadium struggle

“It's not a data center environment, so often our equipment is in various places around the stadium, because of the size of it,” says Banks. “We have to have a lot of local equipment dotted around to supply the whole stadium so we can't just have a centralized point where everything is air-tight and up to the standards that most larger organizations or more office-like organizations would have. 

“You can have a van driving over your fiber cables, or a beer keg being dropped on your data cable. That is the reality of our day-to-day.”

Keeping the stadium connected and ensuring systems remain online is absolutely essential to the smooth running of match days for Fulham FC. Everything from site-wide security to keeping vendors’ payment systems online relies on Banks and Bhandari’s expertise. And, to keep these systems online, the team relies on an arsenal of equipment.

“We have network locations dotted around everywhere,” says Banks. “So it's not like one rack with three switches, and that runs everything. No, it's 20 racks with three switches each and various other pieces of network equipment.”

All equipment is spread between the four stands of the sprawling Thames-adjacent stadium, which makes patching each switch the biggest of their headaches, they say. Adding to the obvious geographic complexity, the pair also need to instigate upgrades in stages to minimize downtime.

Top-flight cyber security

Windows and Linux servers are patched monthly, they say, but often networking equipment is only patched when the team receives “a serious enough bulletin” from vendors such as Cisco. The strategy falls in line with most other organizations, realistically speaking. Of course, in a perfect world, everything would be patched seamlessly and lightning-fast, but with issues arising from applying fixes, it’s common to see patching strategies take a similar form to Fulham’s.

Like any Premier League soccer club, the need for robust cyber security is essential. Fulham were promoted to the top flight for the 2022/23 season, having spent the previous three seasons in the second division. The team had to be even more scrupulous than they were used to. Phishing, for example, is a major concern, given the Premier League’s requirement for club officials’ contact information to be posted online. Attacks, therefore, targeting the CEO and finance director, among other senior staff, are common.

Fulham FC has never actually succumbed to a serious attack, and it’s perhaps due to the team’s round-the-calendar work to achieve a state of cyber readiness that’s contributed to the achievement. When the weekend stars spend time over the summer relaxing in various sun-soaked locations across the world, Banks and Bhandari are commissioning annual penetration tests to strengthen the club’s cyber resolve. The most recent one, Bhandari says, shone a light on the club’s password practices.

“I think the biggest [issue] we had to take quite serious action on was password complexity,” Bhandari says. “Basically purchasing a tool that will restrict password reuse and enforce more complexity onto passwords. 

“There was also a vulnerability that allowed the pen testers to gain access to our domain controllers. But, that was on an authenticated network with credentials, so they were kind of halfway there. 

“But they managed to find one very old unpatched Linux box, which we actually didn't even realize was still running. Then once they actually decrypted our AD passwords, that's where we saw the biggest report. So it was a 100-page report and 99 pages were just bad passwords.”

The club already enforces a three-month password reset, but the pen test’s findings indicated it wasn’t enough. They invested in Specops’ password policy tool to enforce greater complexity beyond the weak passwords that were too often used. Bhandari says the names of the manager and key players, for example, featured heavily in passwords and weren’t the most secure.

A backup strategy for the modern age

Bolstering its cyber security posture even further, Fulham FC recently signed a three-year partnership with Acronis and its technology partner EveryCloud to implement Microsoft 365 backups across the club and the club’s charitable foundation, and disaster recovery for the foundation only. The backup solution for the club’s Microsoft 365 files is already up and running, with the remaining upgrades expected to complete shortly. It’s the key component of Fulham FC’s partnership with Acronis and EveryCloud.

It’s also already proven its worth; in just one month, for example, the team restored a staffer’s lost email files. Fulham’s IT experts declined to detail the nature of the deleted emails but did say such incidents “could have quite significant impacts”, including, say, carrying implications for player contracts.

Agreements made over email are contractual and binding — so if an agreement between two clubs is disputed, and one couldn’t provide the evidence to support its position, it would carry huge ramifications.

“We can't quantify these impacts because we just provide the service to the business — a lot of it is outside what we get told,” says Banks. “So it could have been a dispute over a £5 refund for a ticket, it could have been a dispute of a £70 million player.”

Banks also highlights the importance of having such measures in place at Fulham FC’s foundation, which runs soccer outreach schemes with children and disabled players, among other initiatives. Banks alluded to a child sexual abuse scandal that rocked English soccer in 2021, using it to underline the need to have records of emails and other files when safeguarding vulnerable people.

Backups, in addition to preventing the loss of sensitive files, also act as the club’s ransomware defense. The club stores 30 days’ worth of daily backups, after which time the only restore point is the first day of every month. Beyond that, the club keeps restoration points on a once-yearly basis.

As long as any attack doesn’t impact systems that protect fans’ safety, such as CCTV cameras, radios, and stadium floodlights, a system shutdown could commence. In this scenario, stadium turnstiles would open fully to anyone who approached them and Fulham would simply “assume that everyone who comes in has the tickets”.

If fan-safety systems were to fall, that’s when the Premier League itself would step in, either delaying or postponing the game. Luckily, Fulham FC has never been targeted with ransomware to the extent it’s prevented a match from taking place.

Naturally, Fulham FC stores as much data as any large organization. From fan and member data, to commercial contracts, player performance analytics — all the way to match highlights from the 1900s. As such, the club must be smart about what data is backed up and what isn’t.

“Anything that's nice to keep, but isn't business critical, we potentially don't back up and just keep it on redundant storage,” says Banks. 

“Because the other thing we have to take into account with cloud backups, and the more files we send to Acronis, the more impact it has on our bandwidth. So, obviously, we have to run these backups overnight, and be selective because we haven't got unlimited bandwidth. We have to strike a fine balance between cost, performance, and our requirements.”

Fibre optic 10GbE networks are becoming affordable for SMBs, and TP-Link’s TL-SX3016F switch intends to make them even more appealing. This slimline 1U chassis offers 16 SFP+ 10GbE ports for a shade over £400, to deliver an incredibly low cost per 10GbE port of only £26.

You’ll need to factor in the cost of laser transceivers and fibre optic cabling, but a big advantage over copper 10GbE installations is they support much longer distances. Furthermore, unlike blue-chips that may lock you in with their own coded transceivers, the TL-SX3016F has no restrictions and happily accepted our PlusOptic 10GBase-SR models, which cost £35 each and support connections of up to 300 metres.

The TL-SX3016F has the capacity to handle a high data throughput, as it’s endowed with a speedy 320Gbits/sec backplane. Combine this with a high 238 million packets per second (mpps) forwarding rate and you have a switch that easily has the measure of more expensive 10GbE SMB products like the Zyxel XS1930-12HP.

A feature you wouldn’t normally expect to see at this price is dual redundant power supplies. Two sockets are provided at the rear, with the first PSU acting as the primary source; if that goes down, the second PSU steps in immediately to take over.

You also get a choice of management modes. The switch offers its own web browser interface or it can be integrated into TP-Link’s Omada cloud portal. In standalone mode, the web console sensibly requests the default administrative password to be changed and then presents a dashboard view showing the status of each port.

Tabs are provided for quick access to the various functions, with the switch offering an abundance of Layer 2 (L2) features, including port, MAC and protocol-based VLANs, and QoS traffic prioritisation, plus static and LACP link aggregation groups. VoIP networks are covered as the switch identifies traffic from IP phones using their organisationally unique identifier (OUI) and automatically prioritises it by dynamically creating voice VLANs.

The switch also offers basic Layer 3 routing capabilities. Referred to as “Advanced L2+” or “Layer 3 Lite”, it supports static IPv4 and IPv6 routing, but not the dynamic routing found in more costly full Layer 3 switches.

There are some nice surprises for switch monitoring. Along with support for all versions of SNMP, the switch has an internal sFlow agent for sending packet sampling data to a collector device. We found this very useful as we pointed the agent at a host running the WhatsUp Gold 2022 network-monitoring software, which promptly started reporting back on switch port utilisation, traffic throughout, the top detected apps and all endpoint conversations.

To remotely manage the switch, you’ll need to deploy TP-Link’s cloud controller at each site. TP-Link provides free Windows and Linux software versions, but for testing we use its preconfigured Omada OC300 hardware controller which costs around £110; the smaller OC200 costs only £63.

The cloud portal lists all assigned controllers; selecting one takes you to its own web interface, which provides a central location for managing and monitoring all TP-Link switches, routers, gateways and wireless access points. New devices initially appeared in the controller’s web console as “pending”, and once we had “adopted” the switch, it disabled its local web interface and took all settings from the controller.

The portal provided plenty of information about switch operations but some features, such as L3 options and the sFlow agent, can only be configured when it’s in standalone mode. This wasn’t a problem as we set these up before adopting the switch into Omada and set the portal’s override feature so all local settings were retained.

The TL-SX3016F is a great choice for small businesses that want a high-performance fibre 10GbE core switch and larger organisations looking for a low-cost aggregation solution. It partners a big helping of SFP+ ports with a great range of features, and all at a price the competition will be hard-pushed to match.

TP-Link JetStream TL-SX3016F specifications

Zyxel takes multi-gigabit switching to the next performance level with the deceptively small XS1930-12HP, which is positively bulging with features. Not only are all 12 ports on this desktop-sized unit the 10-gigabit (10GbE) variety, but the first eight provide 802.3bt PoE++ services.

The switch is incredibly versatile. All ten copper ports support 100Mbits/sec, 1GbE, 2.5GbE, 5GbE or 10GbE connections and you can easily see their operating speed as each one has a five-colour status LED. The last two are standard 10GbE SFP+ fibre ports for high-speed uplinks over long distances.

A generous power budget of 375W means the switch can easily provide 30W on all eight ports with enough headroom to power quite a few high-performance Wi-Fi 6 access points (APs) that need up to 60W. The LED status panel comes into play again here as it shows which ports are providing power, while a vertical light bar alongside reveals the total draw.

At its foundation, the XS1930-12HP is a Layer 2 (L2) switch with basic Layer 3 routing capabilities. Sometimes referred to as “Layer 3 Lite” or “Advanced L2+”, it supports static IPv4 and IPv6 routing, although not the dynamic routing found in more expensive full Layer 3 switches.

Standard L2 features are plentiful and include port and protocol-based VLANs, QoS traffic prioritisation plus static and LACP link aggregation groups. VoIP networks are covered since the switch identifies traffic from IP phones using their organisationally unique identifier (OUI) and prioritises it automatically by dynamically creating voice VLANs.

Then there’s management. You can opt for standalone mode or use Zyxel’s Nebula Control Center (NCC) platform to move it into the cloud. The local web console fires up a quick-start wizard that insists you change the default admin password and SNMP community name and, although it doesn’t provide any switch graphics, tables offer plenty of detail on system utilisation plus port and PoE usage.

If you’re concerned about exceeding the switch’s power budget, you can set priorities on individual multi-gig ports. Assign one of three priorities to each one and, if the drain reaches the total power threshold, those with the lowest priorities will be switched off first.

Moving the switch to cloud management is a cinch. We used the Nebula iOS app on an iPad to scan its QR code and assign it to our site. A minute later the switch appeared online in our cloud portal, received a firmware update and had configuration access in its local web console disabled.

Selecting the switch from the portal’s monitoring page presents a full status readout and location map, while a multicoloured graphic below shows all active ports, their connection speeds and those delivering power. Graphs are provided for network traffic and power consumption, but if you want more than a rolling 24 hours of data, you’ll need to upgrade the free NCC basic service to a Plus licence, which extends this to seven days.

The portal provides a range of global settings so any switch added to the site automatically receives them. These include QoS profiles, port mirroring and voice VLANs. Security can be stiffened up by limiting all management access to specific host IP address ranges and only allowing the switch to obtain its management address from authorised DHCP servers.

Individual ports on the switch can be configured from NCC, with options for enabling or disabling PoE services and fixing the connection speed, and a Plus licence enabling bandwidth controls. Up to five global PoE schedules can be created and assigned to selected ports so you can decide when powered devices such as wireless APs are active.

Its smart combination of high-performance multi-gigabit ports and generous PoE++ services puts Zyxel’s XS1930-12HP in a league of its own. SMBs that want to deploy the latest Wi-Fi 6 APs and team them with fast links to storage and servers won’t find a better alternative.

Zyxel XS1930-12HP specifications

Many frustrations and failures in business computing come from things that ought to be simple. The problem is, it’s easy to focus on big, showy, chewy problems and neglect the little things that can have a big effect on overall productivity. With a little planning – and sometimes a little user participation – there’s an awful lot you can achieve in a single day, with minimal disruption to the business and no need to book in expensive consultants or advisers.

Aside from the direct object of a project, there can be a cultural benefit too. IT departments are largely reactive, but a one-day project gives them a chance to demonstrate a bit of proactivity and forward thinking, without completely rewriting their role. Involving the users encourages a bit of self-reliance too – an increasingly desirable trait as hybrid work becomes the norm. Ultimately, the odd one-day project can only be positive for the relationship between IT staff and the rest of the business. Here are some ideas to start you off.

Fire drill day

This is the granddaddy of one-day projects – when you take an afternoon out to pretend there are flames shooting out of the back of the server and spreading around the building. The experience can be both hilarious and chastening, with lessons that go far beyond the IT department as it sinks in exactly how reliant the business is on its flammable assets.

Indeed, in these days of cyber resilience and perilous international relations, it’s worth considering extending the fire drill concept to specific bits of your IT environment. Take a day out to thoroughly investigate the failures and shortfalls that take place when a particular system goes offline and can’t simply be brought back with a remote power cycle.

Log file reading day

I can see the wrinkled noses as I type. Log files? What use are they on a laptop? Actually, they’re just as much use as they were on the Windows 2000 Server where you first saw them. It’s true that most Windows PCs go for years without the Event Viewer ever being opened, but that’s not something an IT department should be proud of – the information contained therein can provide answers to everything from obscure network problems to pinpointing the exact moment power was restored in one of those sneaky overnight outages we’re not supposed to fret over.

It doesn’t require much specialist knowledge, either. Events aren’t mystical things; if anything they’re a bit too explanatory, making the beleaguered sysadmin work laboriously through the verbiage to get to the crucial information. But at the end of the hunt, you’ll often find references to illuminating knowledge-base articles. And you may start seeing patterns in distinct devices: persistent loss of connection to shared network drives? Exchange servers you can’t reconnect to after your machine’s gone to sleep? Such subtle failures become much more obvious when you let the Event Viewer do the searching.

There are a slew of management utilities which will consolidate many machine’s events and log files and go looking for patterns in the mash of data thus created. However, quite a lot of them are integrated into much larger network-management software suites, which with the best will in the world, you’re not going to deploy as a one-day project.

Hunt the temp files day

While I might not want unqualified hands wandering about the file structure of my Exchange server, there’s definitely no harm in getting a bit of user-participation going when it comes to finding and nuking the myriad types and collections of files which you can’t see are being collected by your PC.

Don’t underestimate the potential gains. Some applications generate huge temporary files; others churn out smaller items, but in great numbers. I’ve seen printers create a single file per print job or sheet of paper, and not always cleanly delete them once the print run is over.

It’s not just about disk space. There are a whole raft of different methods of picking up a file in a directory, from the perspective of a developer, and temporary files can have a big impact on performance. This is why things run faster in neglected Windows 7 machines once you clean out all the .txt files in the Windows directory, and why periodically emptying C:\WINDOWS\Temp can seem to turbo-charge certain operations.

There are downloadable utilities which can automate the process of clearing out temporary files, but it doesn’t hurt to engage your staff and help them gain some understanding of these not-too-mysterious secrets of business computers in action.

The other Patch Tuesday: Physical cabling

Are you sure that all the leads and interlinks that make up your physical network are working as they should? Really sure? After all, even the small patch leads used in an equipment rack can get pinched by heavy servers and steel uprights.

This isn’t something you can do in software. Faced with a big pile of patch leads, your best bet might be to employ a Fluke cable tester, though those come with a reassuringly plump price tag. However, many modern switches can report on cable damage via their web interfaces, and there are cable-testing modes hidden away in some of Intel’s network card drivers too. Only the most hardcore techies tend to discover those, not least because they don’t show up when you’re using the default Microsoft drivers.

Checking your cabling makes for an excellent one-day project, because if it gets to 4.59pm and you haven’t quite finished, you should still be left with a stack of known-good cables. Admittedly it’s quite a boring process and not one to leave with the users, but it shouldn’t get in the way of normal operations either – briefly pulling a cable out of a PC and replacing it barely interrupts the session, and both ends ought to resume the link seamlessly.

Drivers and firmware day

I’ve suggested involving non-IT staff in your one-day projects, but this is one I’d recommend you leave to the specialists. Playing with firmware used to be a reasonably safe proposition, because the sorts of device that had firmware used to be ones that did quite simple jobs. Now, though, the firmware has become more complex and the devices themselves have become more critical: even if your router isn’t running the latest firmware, you can’t just shut it down for an update.

To be honest, when dealing with new clients and networks in unknown states of repair, I always try to see whether, rather than performing an in-place upgrade to a vital core bit of the network, we can replace it with something newer and better. That doesn’t always mean consigning that old device to the Cupboard of Shame: quite often it turns out that running two routers in a lightly connected setup gives you an overall faster design, while mitigating the risks of firmware upgrading.

Of course, Wi-Fi routers aren’t the only devices with firmware. Even the humble network switch has a brain, and it’s worth making sure it’s up to date in the name of security and performance. Then, of course, we get into the wonderful world of user hardware: it’s quite incredible how much better the latest Intel Ethernet drivers are, compared to the ones burned into the initial releases of Windows 10. Unless you’re working in a very small office, however, it might be a stretch to get the drivers found, identified, downloaded and matched to all your PCs in a single day. And I’m willing to bet a small pile of actual coins of the realm that even if you can do that, you won’t have time to go round and optimally configure all the extra network parameters and controls provided by the driver.

‘Wooden Spoon’ discovery day

Businesses grow in fits and starts. It’s very unlikely indeed that all the PCs in your office are of the same age, spec, and operating system revision. I’d go so far as to say that the only companies where I’ve seen such standardisation are the ones in which I’ve recommended it, to provide a bit of resilience and redundancy when things start going peculiar.

The Wooden Spoon is of course the award that goes to the oldest, slowest, most decrepit computer that’s still in regular use. For logical but ironic reasons, this will frequently be the one that’s doing the most business-critical job, and being used by the most hard-pressed member of the firm. If you can prise them off it for just a day, the gains can be enormous: I once replaced a Pentium 4 workstation with something that had two six-core Xeons and an onboard RAID card, representing a leap of about six hardware generations in one afternoon. The old machine had been taking about 40 minutes to run the developer’s modelling code; the new one ran the same code so fast we couldn’t press the stopwatch quickly enough.

There’s no need to stop at workstations: printers, network switches and other key components of your business have a tendency to quietly reach retirement age while your attention is focused elsewhere. Tread a little carefully though. You can’t always just whisk away an outmoded item and have a fully functional replacement configured and running by the end of that singular day. Start with a focus on identifying the worst mismatches between hardware and job requirements, and make a realistic plan from there.

Password day

Managing access – to cloud systems, websites, customers and all sorts of other things – has never been more important. Yet the actual management of passwords has become increasingly fluid. In practice, almost every worker nowadays has two sets of passwords – their own, and all the other ones they need to use in the course of doing their job. Password reuse has lamentably become so commonplace.

It’s vitally important, therefore, to find out what passwords everyone’s actually using – along with where they got them from, and who else knows them. Obviously this isn’t something that should be on a cloud-hosted spreadsheet, though: locking a password booklet in your office desk drawer might be smart. It might then even survive a fire drill.

Shadow amnesty day

“Shadow” here refers to shadow IT – the unsanctioned use of third-party services inside a business. It might be Google services, Skype or what have you; plenty of workers, especially the younger generations, are wholly accustomed to using free consumer tools to get things done, and won’t hesitate to fire them up in a professional setting. It could also include any sort of activity that goes outside of your documented and managed processes counts.

The aim of a shadow amnesty day isn’t to cut out all the shadow service uses. Primarily, it’s to identify the areas where your in-house, properly managed IT department aren’t giving users what they need. If you can then use that information to set up a proper sharing platform that means people no longer need to put confidential documents on Dropbox, that’s a win for all concerned.

Advanced project: Duplicates day

This one isn’t all that advanced for techies, but it can feel that way for users when you ask them to go hunting around their systems for duplicate files. Whether you use Windows 10 or Windows 11, the operating system doesn’t make this easy, especially when documents are spread across a variety of storage platforms and security settings.

You might question whether this is a worthwhile use of time and energy. Many people do – many people, that is, who haven’t seen the benefits of de-duplication, as found on the most upmarket NAS and storage server gear. These monsters – a single full-height rack packed with disks qualifies as a small one – can identify data that’s duplicated across multiple files and users, and refer hundreds of thousands of files all to a single block copy. The effect on storage requirements can be breathtaking: I’ve seen the time taken to back up an entire Exchange Server virtual machine (VM) drop from several hours to a matter of minutes.

Okay, so we’re not going to match those benefits by clearing out a load of individual duplicates, but we can still make a significant difference to your storage needs, which will in turn have a knock-on effect on the speed and efficiency of your backups.

The best approach is to standardise on a few lightweight utility tools. The first task is to analyse exactly where your storage is being eaten up, and to identify where you most need to free up some space. Then, you can use one of numerous free tools to search and destroy any space-wasting duplicates you may be harbouring. Personally I split up these operations by size; I start by searching for duplicate files over 500MB and then work my way down. This way, you can achieve big space savings very quickly, and it’s up to you whether you continue to inspect the smaller files.

Cisco has announced two new products that mark a shift in its approach to network management that will shape the company’s portfolio for the coming years.

The networking giant announced Cloud Management for Cisco Catalyst and Cisco Nexus Cloud, two new cloud management services for its campus networking and data centre switch products, respectively.

Cloud Management for Cisco Catalyst will give customers the option to manage select Catalyst switches and wireless devices in the Cisco Meraki dashboard.

The aim of the new service is to provide customers with the option to increase visibility into the performance of their networking hardware and promote easier management through a single application view.

Catalyst customers can also still use Cisco DNA Center for on-prem management if they wish. They can also use a combination of the two if their needs require it, such as network segmentation for security purposes.

“Meraki is undeniably the simplest cloud management platform on the market, the most adopted, the highest deployed, and bringing these two things together: the powerful simplicity without compromises - I truly believe you can change the way IT is managed to carve a new path that delivers sophistication, power, and simplicity,” said Todd Nightingale, EVP and GM at Cisco enterprise networking and cloud.

The goal for Cloud Management for Cisco Catalyst is to make troubleshooting the network much easier, added Nightingale, speaking at Cisco Live 2022, and will provide greater visibility into switch configurations, traffic flows, and connected clients.

Cisco’s Nexus data centre switches will also soon be manageable through the new Cisco Nexus Cloud platform, due for release in Autumn 2022.

The core features of the platform were driven by customer feedback and it will “revolutionise” the way customers manage data centre products and also entire private clouds, said Nightingale.

The platform will be powered by Cisco Intersight, the company’s cloud operations management platform, which will mean customers will be able to more easily manage products across public cloud, private cloud, and edge computing environments of any size or scale, according to Cisco.

“Our customers choose to run their businesses on Cisco technology because we sit at the intersection of networking, security and cloud,” said Nightingale. “We believe the network is the foundation for the modern enterprise and must deliver agility through simplicity. Cisco is addressing our customers' most important concern, which is managing complexity through smart, data-driven platforms that power a digital business.”

With Nexus Cloud, customers can gather telemetry from each device and view the network in a dynamic topology, which provides a view into “every endpoint, every virtual machine, every server, and every switch on the network”.

It also introduces capabilities to monitor energy consumption that will help businesses reach their environmental, social, and governance (ESG) ambitions. Nightingale said even more environmental features will be coming to the platform in the future, including the ability to limit power consumption in certain areas of the network.

Cisco’s philosophy for network innovation

Key themes have emerged throughout Cisco Live 2022 that revealed the company’s priorities in its approach to product innovation, with particular focus being placed on simplification, visibility, and a shift to a cloud-managed approach.

Rebecca Stone, SVP of customer solutions marketing and CMO at Cisco, said that the increasing complexity of enterprise networks across all industries is growing to a point where the technology is becoming “less scalable and less intelligent”.

Evolving consumer demands and different working environments due to hybrid work arrangements, in addition to the mounting standards that need to be facilitated such as Wi-Fi 6E and private 5G, are also contributing to the complexity networking operations teams are struggling to face, she added.

“It’s causing the IT experience to suffer,” said Stone, speaking at a press and analyst event. “[IT teams] can't manage, they can't scale to all of the demands that are coming today, and so the user experience is really suffering as well.”

Stone gave the example of everyday professionals facing increasing amounts of friction signing in to all the different services such as email, VPNs, collaboration tools, and other products - using multiple identity and access management clients to access the tools they need every day.

Cisco said it believes the best way to build a solution that can address these issues is to approach it from a platform perspective, which can help unify technologies and ultimately improve the experience for both IT teams and the end-users too.

Meraki management for Catalyst, and Nexus Cloud, aim to address these issues while adhering to three core principles: cloud-driven automation, simple solutions, and providing deep networking insights through greater network visibility.

Mobile World Congress (MWC) will return to the Fira Gran Via exhibition centre on 28 February in the fullest possible way for the first time since 2019. In the intervening three years, the world – including the tech industry – has changed unfathomably. Beyond mandatory FFP2 face masks and health certificates, however, we’re yet to see how COVID-19 will affect the annual Barcelona-based event.

Two of the biggest talking points at MWC 2022 – 5G and a new flagship Samsung phone – makes it seem as if time has stood still. Only when you scratch the surface does it become clearer how much has actually changed. In early 2019, 5G was only on the cusp of reaching UK smartphone users and businesses. Three years on, though, it’s being used to enable artificial intelligence (AI) and power the metaverse – a term which might as well have been considered alien in pre-pandemic times. Rather than another Galaxy S smartphone, too, which was actually announced at the beginning of February, we can expect a more affordable alternative from the ​​Galaxy A and M ranges, according to Forrester VP and principal analyst, Thomas Husson.

With 5G among the key topics of this year’s event, coverage surrounding 6G is coming off as rather underwhelming. This is especially surprising given the recent developments on that front, including record speed tests and international partnerships. Although MWC’s agenda includes sessions such as the Future of 6G and the Road to 6G, actual concrete announcements for potential 6G-related products, whether on infrastructure or the technology powering it, seem to be missing in action. Should we file a missing cellular network report? Give us until next week to decide.

On the flip side, the future of Wi-Fi seems to be in safe hands. Although the benefits of Wi-Fi 6 are only just starting to be realised, we’ll see some big announcements regarding Wi-Fi 7 and its capabilities. Another major theme will be extended reality (XR), which is likely to steal some attention from its constituent cousins augmented reality (AR) and virtual reality (VR).

In keeping with tradition, meanwhile, the wide-reaching conference will summon the biggest manufacturers from across the world – although the majority of them may hail from Europe. The rise in COVID-19 cases in China might affect travel plans for some of the biggest Chinese tech giants, which have historically been front-and-centre at MWC. To resolve this issue, CCS Insight chief analyst Ben Wood predicts that companies including Oppo, Vivo, and Xiaomi aim to deliver a strong MWC presence by deploying their European-based staff. For one particular Chinese company, Huawei, the return to MWC tastes much different.

The firm has taken a hit like no other in the last two years, facing significant infrastructure restrictions in a number of Western countries, including the UK, and taking a 30% revenue hit in 2021. Huawei might see MWC as a chance for redemption – at least slightly – on European turf, as its focus lies on “surviving” the next five years.

From Wi-Fi 7 to XR, this year’s MWC is set to be a welcome break from covering international tech events from the kitchen table. Remote working is one thing, but missing out on hands-on experience, new friends, and free drinks? Barcelona, here we come.

A slow network hampers your productivity, and it’s also frankly embarrassing – it sends a message to your staff, and to any external parties who may access it. It’s an area, though, in which major improvements can frequently be made with some tweaks to your settings that don’t cost a penny, aside from the engineer time and expertise required to implement them.

Getting the required results isn’t necessarily straightforward, because there are so many ways a network can be slow. One business I shall not name used a cheap Wi-Fi base station and a corporate Dropbox account for storage and collaboration. The network was fast enough for day-to-day work, but not for continually syncing large amounts of data to and from everyone’s computer, resulting in workers having to hang around and wait for all their data to come down the line before they could leave for the day.

There’s a major divide as far as a large proportion of companies are concerned, though, and that’s wired versus wireless networks. There are all sorts of fundamental differences between these two networking models. For example, you could spend thousands installing a top-speed Wi-Fi 6 connection, while a patch cable can be bought using the dusty old pennies in the bottom of the office coffee kitty.

Marketing teams want us to believe pricier Ethernet cables equate with higher quality, just as we’ve been told, for decades in the hi-fi market. There are a litany of cables available for roughly £30, although the quality of most cables is high enough quality to survive even the most benign of office environments. Their main adversary is the office chair castor wheel – a device capable of delivering hundreds of kilos of crushing force per square inch. I recommend you think of an Ethernet cable as a consumable item, like a toner cartridge or light bulb. Once it's crushed, chewed, stretched, kinked or cut, that's the end of it. Not too long ago I ran a cable tester over all of a client’s leads and found 60% were degraded in some way – which can lead to all sorts of intermittent performance problems.

In fact, if you’re trying to tune up your network, a brilliant first diagnostic is to temporarily replace your wireless connections with a cable run. You might have to fiddle around because not every router configuration treats its Ethernet cable ports in the same way as its wireless clients. It's worth the effort, though, because your basic cable gives you an immediate indication of whether your speeds are being held back by the Wi-Fi, or by something deeper in the infrastructure.

If your wireless connection does turn out to be sucking the performance out of your network, there are ways to tune up your Wi-Fi, maximise its speeds and reduce interference. Often, though, the real solution is a hardware upgrade: that’s sadly unlikely to count as almost free, but it could still pay for itself in time if it lets staff work more efficiently.

A matter of protocol

In almost any network, wired or wireless, there are certain structures that you can tune for performance. It might surprise you how similar they are across both Ethernet and Wi-Fi. It's called Ethernet because, even though data doesn’t travel through the “ether”, the concept was originally based somewhat on running radio transmissions over a cable. When Wi-Fi came along, it wasn’t a problem to adopt the same philosophies and terminologies.

Setting an appropriate maximum size for your data packets, for example, can enhance efficiency, whether those packets are inside an eight-core copper cable or floating through the air. There's a lot of fossil knowledge that says “leave the MTU and packet size alone”, because back in 1994 many routers weren’t able to dynamically switch packet size according to the settings for each class of connection. This is no longer 1994, however, and there’s now more CPU power inside a network card than you might have had in an entire computer. We no longer need to cling to the idea that cheapness and simplicity are virtues in network traffic infrastructure; we’re finally free to take full advantage of all the features that were envisaged when the standard was originally drawn up.

Turning on Jumbo Frames in your PC's Ethernet card settings isn't a magic-bullet hack, though, and you shouldn’t expect a blinding flash of power. Rather, it should give you a small increment of speedup, and turning on full duplex transmission and flow control – while perhaps turning off energy-saving features – should do the same, adding up to a worthwhile performance improvement. Of course, these features need to be supported and enabled at both ends of the cable – which brings us into the daunting but important world of network switches.

What's in a name? Switches decoded

Is your switch managed, or not? In band, or out? Can you run cable diagnostics from it? What about OSPF, trunking, and fibre support? The language of network switches seems to be designed to be discouraging, which is a shame because having the right one, and configuring it correctly, is key to running a fast and secure business network.

It doesn’t have to be all that complicated. A managed switch is one that lets you set those key tuning parameters at the switch end, so the PC doesn't have to be in the right mood to get the right negotiation of link speed and protocol. There’s also a spread of hardware and configurations to allow for the broad range of different business interests reliant on network switching. Some companies have to record every single packet, in case it turns out to be important in an international market, options trading, or high-powered finance environment. Others do a lot of their customer work on the phone, so they opt for a VoIP-friendly switch, which will have quite different design priorities to those made for hosting centres or networks where the majority of the traffic is between virtual machines.

Then we get into layers. You might well have heard of Layer 2 switches versus Layer 3: these designations refer to the ancient but still vitally important ISO seven-layer networking model. Layer 2 is machine addresses, so a Layer 2 switch lets you divide up your LAN into VLANS, whose traffic mostly does not need to intermingle. A Layer 3 switch goes a stage further by managing IP addresses for VLANs and allowing a framework of routes and little cul-de-sacs of machines to be built out of each other's way. There are Layer 4 switches, too, which analyse the traffic itself and make realtime routing and priority decisions – but if you’re in a business that requires that sort of technology, you probably already know all about it.

If performance is your priority, then it may be a mistake to pursue complex topologies anyway. The more complicated the network environment, the more slowly it tends to run. If you mix up a hundred VLANs and pass them all down a VPN tunnel, don't be shocked when the whole thing just about grumbles up to 256kbits/sec.

Instead, you can think about simplifying and minimising the amount of traffic that’s exchanged. One recent IT buzzphrase is zero trust, which is the idea that every object, device and cable on your network should have an identified device that uses it, and nothing else, connected to a nominated switch port and with a list of traffic types it may accept. This may sound radical but it can be done pretty cheaply – quite a lot of it is executing a design, typing in a bunch of addresses and port ranges and so on. Almost any decent Layer 3 switch could be used to set up a network like this, and once you chop out the chattering nonsense, the traffic you don't normally bother to identify and cut out, then your network is quite likely to run better.

The catch is in the upkeep. Making an initial zero trust design isn't too hard, but you have to grow and maintain it every time something changes on your network. For every new service or device on your network you have to decide what access is legitimate. This is where thinking architecturally pays off. Given a single L3 switch, you can have several subnets, some of which are zero trust – the ones with the users on – and others which are unfiltered within their own boundaries. If you don’t already have an L3 switch then this might not exactly be a free upgrade, but you don’t need the latest and greatest hardware.

Going IPv6

IPv6 has had pretty poor PR within the business networking world. It’s major advantage is that each device gets its own globally unique address; this makes the burden on big continent-scale routers considerably lower, but for most businesses that’s either irrelevant or a security concrn. The standing advice when network tuning or product testing has been to turn off IPv6, debug everything in v4, and only re-enable IPv6 later if you need to.

There was a lot of sense to that outlook when IPv6 was relatively new. The default everything-on setting was fine for standard components, but it’d trip up badly behaved installs if the network layer didn’t fully implement IPv6 as expected. We must give thanks to those who have spent the intervening years debugging their code, because nowadays it all seems to work pretty well: your machines should all be happily able to pick up link-local addresses and reach one another through the IPv6 address space without any particular need to fall back on IPv4.

The remarkable thing is, because of the way IPv6 can be encapsulated over IPv4, you don’t even require a dedicated IPv6-capable network switch to make it all work. If you want to connect to the internet over an all-IPv6 connection, that’s possible too, either through a VPN or modern hardware and a cooperative ISP.

On its own, switching to IPv6 will make almost no difference to your internal network speed, though. The major boost comes if you manage to line up all the tweaks mentioned so far. If you can turn on jumbo frames, and flow control, and prevent the switches from trying to negotiate the wrong duplex setting, then IPv6 is the icing on the cake – at least, for operations that run in a standard Windows-supported way.

10GbE and faster

If you want to ensure your data is whizzing around the building at top speed, then 10GbE is the standard to aim for. 40GbE and 100GbE standards do exist, but the hardware is prohibitively expensive, and the interconnects aren’t cheap either. Meanwhile, a 10GbE card can be easily had for under £100, and it should be at least partly operational with your existing wiring.

You’ll still need to budget for a 10GbE switch, but something like the Netgear Pro GS110TPv3 can be had for a tad under £200. This is a decent basis for performance experimentation; a lot of supposedly 10GbE devices will immediately fall down to lower connection speeds if your configuration isn’t precisely to their liking. You need to be sure you’re getting the performance you thought you were paying for, so go through all the checks I’ve mentioned, and make sure you apply any required software tweaks to make your various machines use the infrastructure properly. Moving to 10GbE is by no means a drop-in upgrade.

Frustratingly, the spread of support for 10GbE has been slowest in exactly the market you’d most want it – cheap, large-capacity storage devices. Generally, only the bigger NAS appliances (those with eight drive bays or more) are currently offered with integrated 10GbE ports. As hard disk sizes continue to grow, however, we’re at the point where it’s perfectly feasible to deploy a two-drive NAS box with 36TB of storage; hopefully the market will soon catch up to a point where it’s possible to access all that data at the full speed the drives are capable of.

When it’s time to move your office network onto Wi-Fi 6, you don’t want to be held back by a last-generation gigabit switch. To reap the full bandwidth of 802.11ax you need a multi-gigabit back-end. Thankfully, that doesn’t have to mean a jacked-up price: for a shockingly reasonable £263, TP-Link’s TL-SG3210XHP-M2 desktop/rack switch offers eight 2.5GbE ports and two high-speed 10GbE SFP+ fibre uplink ports.

That’s not all. Each of the switch’s eight copper ports presents 802.3af/at PoE+, with a generous total power budget of 240W. This means that the switch can deliver 30W on every multi-gig port at once, making it a great choice for anyone wanting to upgrade or install an office full of APs with a minimum of cabling.

The TL-SG3210XHP-M2 can be managed locally from a smartly designed web console. Standard L2 features are plentiful and include port, MAC and protocol-based VLANs, QoS traffic prioritisation and static and LACP link aggregation groups. There’s even a basic set of Layer 3 routing capabilities – sometimes referred to as Layer 3 Lite or Advanced L2+ – although the switch lacks the dynamic routing capabilities you’ll find on more expensive full Layer 3 switches.

VoIP networks are particularly well served. The switch can identify traffic from IP phones using their organisationally unique identifier (OUI) and prioritise it by dynamically creating voice VLANs. Switch monitoring and diagnostics features are a cut above the rest too: as well as supporting all versions of SNMP, the switch can run an internal sFlow agent for sending packet sampling data to a collector device – a feature we very rarely see at this sort of price.

As for power management, you can assign a priority to each port, and if the total power drain reaches the threshold, devices with the lowest priorities will be switched off first. It’s a simple, practical approach that we’ve seen on plenty of other switches – the only difference being that, in this case, the TL-SG3210XHP-M2’s big power budget means most offices won’t ever trigger it.

For larger organisations the switch can also be cloud-managed. Here TP-Link takes a similar approach to Ubiquiti Networks, with each site requiring a cloud controller to manage all local devices. You can download software for Windows and Linux that will handle this for free, but for testing we used TP-Link’s Omada OC300 hardware controller, a little black box that is preconfigured for the role and costs around £120.

All controllers assigned to your account appear in the cloud portal, and selecting one transports you to its management console. Adding the switch to our site was simple: as soon as we’d connected it to the office network, the controller detected it and presented it in the console as “pending”. We could then click to adopt it, after which the switch disabled its own web interface and took its settings from the controller.

Selecting the switch from the controller’s device page brings up a handy graphic showing which ports are active, with details of connection speed and power status. An overview below shows the power budget, with tabs provided for assigning profiles to each port, rebooting powered devices and viewing attached clients.

One slight oddity we noticed is that not all of the switch’s settings can be accessed from the cloud portal. But you can configure them in standalone mode, and then apply a cloud profile that tells the switch to keep its existing settings after adoption. Another option is to use the profile override feature to manually apply settings for functions such as port mirroring, PoE modes and rate limiting.

For monitoring on the move, TP-Link provides free mobile apps, and we tried using the iOS version on an iPad to check on our sites. We were happy to see plenty of status details, including full power consumption information, and controls to reboot the switch or upgrade its firmware.

Don’t be fooled by the low price: TP-Link’s TL-SG3210XHP-M2 offers a remarkable range of features. With ten multi-gig sockets, eight of them offering full PoE+ services, and a generous total power budget, it’s a terrifically versatile switch, and is particularly well suited to SMBs looking to deploy high-performance Wi-Fi 6 services.

TP-Link TL-SG3210XHP-M2 specifications

DrayTek has a knack for putting together feature-rich network solutions at SMB-friendly prices. The VigorSwitch P2100 is a case in point: this ten-port Gigabit Ethernet switch delivers PoE+ services on all copper ports, with full Layer 2 switching, basic Layer 3 routing capabilities and flexible management options. All for £131.

The P2100’s various connectors are clearly accessible across the front panel, with eight copper GbE ports each capable of supplying up to 30W of 802.3at PoE+ power. The total power budget of 140W may sound small, but it’s more than you’ll get from most other manufacturers at this price. It’s also less restrictive than you might fear: we tried hooking up four D-Link IP cameras to the switch, along with a Linksys wireless access point, and recorded a total power draw of only 13.5W.

The regular network ports are accompanied by a pair of Gigabit SFP ports. With many entry-level switches, using the SFP sockets puts a copper port out of service, but here they’re fully independent, so you can use all ten connectors at once.

It’s not just the connector count that’s generous – DrayTek also offers plenty of management models. For local administration you can use the P2100’s own web console, manage it from any DrayTek router that supports central management, or install the free VigorConnect app for Windows and Linux, which provides on-site monitoring and management of a whole swathe of VigorSwitch and Vigor AP devices. Alternatively, DrayTek’s cloud-hosted VigorACS 2 service offers remote management for around £8 per node per year – and for those who prefer not to rely on the cloud, DrayTek offers the VigorACS 3 version that can be hosted on-site.

For small businesses, the native web console may be fine. It’s very informative, with a graphical home page showing the connection speed of each port, and lightning bolt icons to indicate power. If you’re concerned about the power budget, you can set each port to one of three priority levels – if you run out of watts, the ports with the lowest priority will be automatically switched off first.

Since the P2100 runs the same core firmware as DrayTek’s enterprise switches, you don’t have to dig deep to find network features in abundance. Standard L2 features include port, MAC and protocol-based VLANs, QoS traffic prioritisation and static and LACP link aggregation groups. Basic L3 services (or L2+, as they’re sometimes called) include static IPv4 and IPv6 routing and inter-VLAN local routing – although you won’t get the dynamic routing found in a full L3 switch.

The P2100 pairs particularly well with IP cameras. It identified our D-Link units right away and added quick-access icons next to the relevant ports. We could then use the switch’s ONVIF feature to check on live video streams directly within the web interface.

The switch can even help keep your cameras up and running, by automatically detecting when one has become unresponsive and cycling the power. Optionally, it will capture half-hourly snapshots from selected cameras, and when one goes offline it will send you an email alert with the last snapshot attached.

In addition to the native console, we tried out DrayTek’s central management using a Vigor 2927Lac router; this quickly discovered the switch and generated a hierarchical view showing which port the switch was attached to, along with a simple port status graphic. You can open a detailed view that again shows all switch ports, attached devices and power draw, with optional profiles to control selected ports, apply bandwidth limits and create VLANs.

If you want to power and manage a moderate number of IP cameras, the VigorSwitch P2100 is an excellent choice. Even if you don’t, it makes a very appealing general-purpose PoE+ switch: it’s strong on networking features, easy to manage and competitively priced.

DrayTek VigorSwitch P2100 specifications

Qnap's Guardian series offers an innovative solution that combines a fully-functional NAS appliance with a 16 port, PoE-enabled, fully managed Gigabit switch in a single unit. They're versatile, space-saving solutions designed to provide a wide range of storage and virtualized security services with a sharp focus on video surveillance.

When we reviewed the QGD-1600P 1U rack model, we observed that its minimal internal storage was a potential limitation for video surveillance, but the QGD-3014-16PT resolves this with a big boost in capacity. This makes it more appealing as a high-volume video recording vault and Qnap's range of QVR Pro applications provide a wealth of monitoring services.

Qnap QGD-3014-16PT review: Hardware features

The QGD-3014-16PT is powered by a 2GHz quad-core Intel Celeron J4125 CPU and 8GB of DDR4 memory. The motherboard has two SO-DIMM slots both populated with 4GB modules, which is the maximum the CPU can support – a shame, as this may present some limitations.

At the back, you'll find sixteen Gigabit ports which support 802.3at PoE+ and can each provide up to 30W. The last two copper ports are dual personality and share their backplane links with the two non-PoE Gigabit SFP fibre ports alongside.

NAS storage is handled by four hot-swap LFF drive bays behind the removable plastic cover at the front. Caching is a possibility thanks to the two internal M.2 SATA SSD slots but 10GbE is off the menu as, unlike the QGD-1600P, it doesn't have any expansion slots.

The NAS appliance and switch components are powered independently, which is a smart feature meaning that when a QTS software update requires a reboot, it won't affect devices connected to the switch and drawing power from it. Multi-Gig also comes into the equation, as the appliance has two separate 2.5GbE ports which can be used for management or connecting high performance hosts.

Qnap QGD-3014-16PT review: NAS and switch management

Run Qnap's QFinder discovery app and it'll come back with two devices, each with their own IP address. One is for general NAS management using the QTS web console and other is for the switch's QSS Management web interface, which allows it to be independently monitored and controlled.

The switch can also be managed from within QTS using the QuNetSwitch app which presents an identical console to QSS Management. The home page offers a complete status overview; the ports have colour-coded icons showing their connection speed while those providing power get an orange lightning bolt overlay.

Graphs to the side reveal switch CPU utilisation and temperature, with another below showing overall PoE consumption along with total used and available power. The switch has a comparatively modest power budget of 140W but you can apply one of three power priorities to each port so if the total draw reaches this threshold, those with the lowest priority will be automatically switched off first.

A handy feature is the option to apply PoE schedules to single or multiple ports that define the days they are active and when power is to be supplied. The switch also supports all the Layer 2 features you'd expect to see including VLANs, ACLs (access control lists), port and VLAN QoS (quality of service) and link aggregation.

Qnap QGD-3014-16PT review: 2.5GbE performance

For NAS performance testing, we configured four 10TB Seagate IronWolf NAS drives in a RAID5 array and used the appliance's 2.5GbE ports to hook it up to a Netgear MS510TX multi-Gig switch. We called up the lab's Dell EMC PowerEdge T640 Xeon Scalable tower server running Windows Server 2019 for host duties and connected it to the switch's 10GbE port.

With a NAS share mapped to the server, we saw Iometer report solid sequential read and write rates both of 2.3Gbits/sec. These translated to top real world speeds, with our 25GB drag and drop file copies also returning the same averages.

Backup performance was reasonable - a 22.4GB folder and 10,500 small files were secured to the share at 1Gbits/sec. The Celeron's integral AES-NI engine is very efficient, and copying our 25GB test file to an encrypted share returned a respectable 1.8Gbits/sec with CPU utilization hovering around the 60% mark.

IP SAN performance is equally good with a 500GB target returning Iometer read and write rates of 2.3Gbits/sec and 2.2Gbits/sec. With a dual 2.5GbE MPIO iSCSI link to the target, we saw read and write performance increase to 4.5Gbits/sec and 3.8Gbits/sec.

Qnap QGD-3014-16PT review: Surveillance apps

Qnap provides plenty of surveillance apps. The free version of QVR Pro includes an eight-channel license, and supports 14 days of video playback. You can purchase extension licenses up to a maximum of 128 channels, with the QVR Pro Gold license costing £307 and providing eight extra channels and unlimited playback.

The free version offers plenty of features and its auto-discover feature found our IP cameras without any problems. You can create dedicated recording vaults on the appliance, set up rules to manage motion detection events and use the QVR Pro Windows client to view live feeds on your desktop.

The appliance's HDMI 2.0 ports add extra dimensions as you can connect a mouse, keyboard plus HD TV, directly access the HybridDesk Station app and use its QVR Pro Client app to display all camera feeds locally. Its interface is identical to the Windows client where you select cameras to view in the main pane, create a video wall with multiple cameras, remotely control PTZ functions and play back recordings from your vault.

Video analytics come into the picture with the QVR Face Tiger app, which is available on a monthly subscription with one analytics task and 10 face profiles costing around £2 per month. It's easy enough to use, but we did find that running this alongside all the other QVR Pro apps pushed average CPU utilization as high as 50% and swallowed 40% of available memory.

Qnap QGD-3014-16PT review: Verdict

The QGD-3014-16PT delivers a versatile combination of fully-featured NAS appliance and 16-port PoE+ Gigabit switch that's a lot cheaper than buying separate components. It's an ideal choice as an all-in-one video surveillance solution for remote offices and SMBs - although the maximum 8GB of memory may limit its ability to run extra apps alongside these services.

Qnap QGD-3014-16PT specifications

The EnGenius ECS2512FP is a fairly compact network switch, measuring a modest 330mm across. Don’t judge it on size alone, though, because it packs serious performance into its diminutive chassis. Eight front-facing Ethernet ports all support both regular Gigabit and 2.5Gbits/sec connections, while four 10GbE SFP+ ports allow for high-speed fibre connections.

Even better, all eight multi-Gigabit ports support the 802.3bt PoE++ specification, with an overall power budget of 240W. This means the ECS2512FP can deliver 30W on every port at once, or 60W on any four of them – which should be music to the ears of IT managers looking to deploy high-performance Wi-Fi 6 APs over existing Cat5e cabling.

Three management methods are on offer: you can use the switch’s own built-in web console, integrate it into EnGenius’ free cloud portal or use EnGenius’ ezMaster management software, which will run in a VM, on an EnGenius SkyKey appliance or in the cloud on Amazon AWS. There’s also a mobile app, but this is for monitoring only and doesn’t offer any configuration options.

Standalone management is the least exciting option, but the web console gives you full control of a wide range of standard L2 services, including port and protocol-based VLANs, QoS traffic prioritisation, static or LACP link aggregation groups and port mirroring.

VoIP networks are covered too: the switch automatically identifies traffic from IP phones using their OUI (organisationally unique identifier) and prioritises it by dynamically creating voice VLANs. And if you’re worried about the power budget, you can set each of the ports to one of four priorities: should the total draw threaten to exceed the 240W limit, ports with the lowest priority will be automatically switched off to ensure high-priority devices keep working.

Activating cloud management is as simple as registering the ECS2512FP with your EnGenius Cloud account. We used the Cloud To-Go iOS app to scan in the QR code on the base of the unit, then selected the organisation and network we wanted to add it to. In less than a minute, it popped up ready for action.

So far, so good – although once you’re into the cloud portal you’ll find that it’s clearly been designed with a focus on managing wireless networks. There’s lots of good information on display about APs and wireless clients, but it’s a shame that you can’t tweak the homepage to add selected details about the switch.

Still, on the switch’s own page you’ll find plenty of readouts and controls. Coloured icons give an instant overview of the state of each port, along with its connection speed. Charts below reveal total PoE usage against the switch’s power budget, along with the individual power draw of each port.

From here you can also choose to enable or disable the voice VLAN, add custom OUIs, configure QoS scheduling plus jumbo frames and block access to the switch’s local web console. All port settings can be configured remotely too, as can port mirroring and link aggregation.

If you prefer to use the ezMaster software, it’s quick to set up: it took us only five minutes to install on our VMware ESXi host, and as soon as we opened it up, it had spotted the switch and prompted us to approve it. With the ECS2512FP added to our ezMaster project we could now change its port settings, monitor used and available power, check the status of connected devices and modify their power priorities. You won’t find as many configuration options as the cloud portal offers, but you can launch the switch’s own web console directly from the ezMaster interface. Just note that once the switch has been claimed by ezMaster, its individual port and power settings become unavailable. We also noticed that when devices connected over 2.5GbE, the local console correctly showed their port speed, while ezMaster thought they were down – a minor bug that will hopefully soon be squashed.

As Wi-Fi 6 APs start to become the norm, manufacturers have been curiously slow to offer switches that provide a suitable combination of performance and power. EnGenius’ ECS2512FP hits the target bang on, delivering exactly what you need to start rolling out next-gen Wi-Fi – and does so at a good price with plenty of management choices.

EnGenius ECS2512FP specifications

Zyxel’s GS1350 series of Layer 2 switches make perfectly fine general-purpose PoE switches, but they’re especially well-suited to driving IP cameras. The distinctive Auto Recovery feature keeps a watchful eye on all powered devices and can automatically power-cycle any that become unresponsive, while Zyxel’s patented surge protection circuitry also protects exterior cameras from lightning strikes.

The switch family comprises four models, ranging from the six-port GS1350-6HP up to the rackmount, 26-port GS1350-26HP. We tested the GS1350-12HP, which presents eight Gigabit Ethernet ports providing 802.3at PoE+ services, along with two pairs of unpowered copper and fibre dual-personality ports.

The unit has a reasonable power budget of 130W, and an extended-range mode that overcomes the 100m PoE cable length limitation, allowing powered devices to be placed up to 250m away. It can supply up to 15W of power in this mode, but it’s not suitable for devices that handle high volumes of traffic as – like most other PoE extenders – the link speed drops to 10Mbits/sec.

For management, you can use either the switch’s own web portal or the free Zyxel One Network (ZON) Windows utility. Root around the web console and you’ll also find the option to enable discovery by Zyxel’s Nebula Control Center cloud management platform: this is currently non-functional, but Zyxel plans to activate the feature in a few months.

The first time you access the web console, you’ll be greeted by a quick-start wizard that insists you change the default admin password and SNMP community name. Once you’re in, you might be a little disappointed to find that there’s no graphical overview of ports in use, but the console does provide plenty of information on switch utilisation, plus details of port and PoE usage; you can keep a close eye on individual power being supplied to each port, see the total draw and monitor power budget usage.

If you’re worried about breaching the power budget, each PoE port can be set to one of three priority levels, with lower-priority ports being automatically turned off should the drain reach the power threshold. You can also set up profiles to control when power is supplied to selected ports, using custom time schedules to automatically disable them when they’re not needed.

Then there’s the brilliant Auto Recovery feature, which uses either LLDP or Ping to verify that the device attached to a specified port is active, and cycles the power if it doesn’t receive a response after a set number of attempts. You can decide how often each port is polled, set the wait time between power cycles and tell the switch how many times it should turn the device off and on again before giving up. You can also generate an SNMP trap and log whenever an unresponsive device is rebooted, although email alerting isn’t supported.

Lastly there’s the Neighbor web page, which shows the status of all powered devices, along with the IP addresses and hostnames of those that support LLDP. Coloured icons indicate whether each device is up, down or being rebooted, and you can manually force a power cycle on any port. If the device is an IP camera, you can open its web interface directly from the switch’s console simply by clicking on its IP address if displayed. Helpfully, if the switch itself needs to be rebooted – for example, for a firmware upgrade – this can be done without cutting the power and network connectivity for attached devices.

If your physical security is being compromised by flaky IP cameras, the GS1350-12HP could be the answer to your prayers, helping to ensure around-the-clock surveillance with no human intervention. It’s also good value, very easy to deploy and more than capable of regular switch duties, making it a top choice for small businesses.

The rapid growth of mobility, personal devices and cloud-centric applications today are driving the demand for bandwidth to an all-time high. New faster wireless standards are now giving us access points capable of gigabit+ speeds. But can your wired network handle the extra load?

If not, you most likely will need to plan for an upgrade to your wired network. However, determining which switches will best meet your needs can be a challenge unless you know what variables to consider.

This guide sets out the tips and information that will help guide you in the right direction.

Over the years, we’ve seen all sorts of innovations in the NAS appliance market, but the Guardian QGD-1600P is something new. This 1U rack-mounter not only delivers all the goodness of Qnap’s QTS NAS software, it also serves as a 16-port managed PoE switch.

This makes it the ideal all-in-one platform for services such as IP surveillance, network security and wireless LAN management. Even better, the first four Gigabit Ethernet ports pump out a full 60W of PoE++, so it’s ready to handle the latest Wi-Fi 6 access points and PTZ IP cameras.

Thanks to Qnap’s QTS software, you’re not limited to basic NAS services, either. The QGD-1600P can run Qnap’s Container Station and Virtualisation Station apps, allowing it to host virtual machines for tools such as the open-source OpenWrt router, the pfSense firewall and Ubiquiti’s UniFi Network Controller software. There’s even an HDMI port at the front for outputting to a local monitor – potentially handy if you want to take advantage of the slick Surveillance Station app.

Internally, it’s all powered by a 1.8GHz Intel Celeron J4115 CPU, partnered with 8GB of DDR4 memory (a 4GB model is also available). The only real limitation is that there’s only space inside for two SFF SATA drives, but you don’t need more than this to take advantage of all the features of QTS, and a pair of PCIe slots allows for further expansion.

Alongside NAS functions, you can also access the QuNetSwitch app, which lets you manage the switch from the same web console. Don’t get mixed up, though: the NAS and switch hardware components are independent, with the switch having its own IP address and a separate QSS Management web interface. Even if you need to reboot the NAS, the switch will stay up, so your network devices will remain connected and powered.

Layer 2 switching features are extensive and include support for port and VLAN QoS, VLANs, ACLs, port mirroring, link aggregation and more. The switch has a reasonable 370W power budget, and the console gives plenty of graphical information about total power draw, with a bar graph showing power usage for each port. If you’re getting close to your power limit, you can assign one of three priorities to each port, and the lowest priority ports will be automatically turned off first if the draw reaches this threshold. PoE schedules are also supported, so you can power down individual ports for specific periods on selected days.

NAS operations are suitably speedy. For testing, we installed a pair of 1.92TB Micron 5200 Max SATA SSDs and used the QTS Storage Manager to configure them as a mirrored storage pool. We then mapped a share on a Dell T640 Windows Server 2019 and saw sequential read and write rates of 113MB/sec and 112MB/sec respectively over Gigabit Ethernet. Real-world speeds are up there, too, with drag-and-drop copies of a 25GB test file averaging 112MB/sec both up and down, while our data backup test completed at a creditable 82MB/sec.

We also tried installing Ubiquiti’s UniFi Network Controller (which we use in the lab to manage our UniFi NanoHD APs) on the QGD-1600P. This proved to be a cinch: we downloaded the pre-prepared Docker image and in five minutes the Linux UNC version was up and running in a Container Station VM. From here we signed in to our Ubiquiti cloud portal account, and once we’d connected our NanoHD APs to the switch, we were able to adopt them and manage our wireless networks in exactly the same way as when we use our Ubiquiti Cloud Key Gen2 Plus UNC appliance.

Clearly, there are NAS options with greater internal storage capacity, but the QGD-1600P is probably the most versatile appliance we’ve seen. It offers the convenience of a NAS and a managed PoE switch in a single low-profile chassis for a much lower price than separate components – and Qnap’s feature-rich QTS software allows it to run a remarkable range of business applications.

Qnap Guardian QGD-1600P specifications

A microcontroller that sits inside a device to control a specific function within it is what we know as an embedded system. This could be, for example, your home’s central heating which is considered to be a type of embedded system.

Embedded systems feature prominently in consumer products and many household machines like toasters, washing machines and microwaves, but they are an essential part of most modern technologies.

Usually, embedded devices are not programmable as they are often designed with a single function in mind. However, depending on what the device is, the software can be upgraded. For example, fitness trackers can be upgraded by connecting them to a laptop or PC.

Because of this, embedded systems must be reliable, since a fault will probably result in the failure of a wider system function or an app, and fixing this can be a challenge.

Hardware foundations of an embedded system

Even though embedded systems used to be based on simple microprocessors, modern ones are usually designed from microcontrollers that come with a specific amount of built-in memory. The difference is that microprocessors only contain a central processing unit, which means that ROM and RAM must be added externally.

These systems appear in a number of forms, with some being stand-alone systems that don’t even have a host, like a video games console. They run certain tasks to a fixed schedule and are a real-time embedded system. A “Network-embedded” system is another example which, as you can probably guess from the name, is a device with a network connection like a mobile phone.

Key features of an embedded system

Embedded systems are typically designed to perform a single repeated function, although it’s true that some can be designed to control the entirety of an operating system. However, regardless of the function involved, they will very rarely be required to do anything more than this task – this makes it an exceptionally reliable component.

They’re described as ‘embedded’ because the component is fixed, and is critical to the overall operation of the system. Those that aren’t critical are described as modular, and can be swapped in and out to allow for new functionality.

Embedded systems are also characterised by their reactive nature. They communicate entirely through sensors or actuators, and if the right response isn’t provided in real-time, the response is considered incorrect and they will not function.

Examples of embedded systems

Examples of embedded systems aren’t hard to find – chances are that you interact with at least a few of such devices on a daily basis. This is because embedded systems are seen in an overwhelming majority of consumer products, from something as small as your Fitbit watch, to your home’s central heating system.

Imagine a typical morning: you are woken up by your alarm clock and make your way to the kitchen, where you set your smart coffee machine to make you an espresso and throw some clothes in the washing machine. You take your car to work, and check the best route according to your GPS navigation system. During lunch break, you heat up your meal in the office microwave. An electronic calculator helps you work out some numbers which you need for a work report, which is in turn sent to your boss thanks to the Wi-Fi provided by your office router. At the end of the day, you check your watch to see whether you can get away with leaving five minutes early. On the way home, you stop by the gym, where your fitness tracker helps you determine your heart rate and the number of calories burnt during your workout. At home, you find some time at the end of the day to play video games on your console because, apparently, it can help level up your career.

All of the devices mentioned above are examples of embedded systems in everyday life. Although they might have different uses and are often sold in different stores, the mechanics behind them are actually pretty similar.

What are the benefits of an embedded system?

Embedded systems usually only have a single function, which means they are able to operate using very little power. They’re also usually very small, and can be crammed in alongside other components relatively easily. Combine all of this with the fact they’re relatively cheap, embedded systems are a hugely efficient means of controlling devices.

As you might have guessed, embedded systems are also incredibly low maintenance, and rarely require direct management, whether that’s changes at the hardware level or in programming.

A component that’s incredibly small, cheap, easy to maintain, and fantastic at doing a single task repeatedly, is the perfect fit for any ‘fire and forget’ devices – those that are required to operate with little fuss and intervention. A handy example of this are the entertainment systems in passenger planes, which were able to function using Windows XP for far longer than a commercial laptop.

What are the downsides of an embedded system?

Despite their invaluable benefits, embedded systems also come with some disadvantages that you should be aware of prior to investing in them. In that way, you can eliminate the element of surprise if something goes wrong, and even opt for a backup plan in case your business is reliant on an embedded system.

For one thing, it’s important to know that embedded systems tend to be difficult to upgrade to a new software or even fix in the event of a malfunction. This is due to the ‘embedded’ part of their nature, which translates to crucial parts being situated deep within the overall machine, and even the smallest change will massively impact the rest of the system.

This means that embedded systems often are hard to successfully debug or fix, which often makes it necessary to deconstruct the entire device, removing the majority of components just to replace one part. In fact, the process can be so convoluted and tiresome that it’s often simpler and cheaper to simply replace the entire machine, which can still entail higher costs than anticipated for something which originally seemed like a ‘quick fix’.

However, the embedded systems’ component interdependability also means that, in some cases, tweaks can be applied to other, more accessible parts, which will then bring the device back to life or original purpose. This is why it’s important to have an understanding of how your embedded system works, as well as a trusted specialist who will be able to recognise the root cause of the problem.

Netgear's ProSafe GS752TPP is designed for SMEs that want more power to their network. The growth of wireless networks, VoIP and IP surveillance (not to mention IoT devices and smart PoE lighting) means legacy PoE switches are fast running out of steam.

Stepping up to the top of Netgear's Smart Managed Pro switch family, the GS752TPP provides 48 copper Gigabit ports and its generous 760W power budget means it can feed 15.4W of PoE to them all. Furthermore, it has enough headroom to support a large number of PoE+ devices such as high-power Wave 2 wireless APs and HD IP cameras.

Along with the 48 copper ports, the switch also has four Gigabit SFP ports. Aimed at providing fault tolerant uplinks over fibre to the network backbone, these are dedicated ports and can be used alongside the copper ports.

The GS752TPP is a Layer 2 switch with basic Layer 3 routing capabilities. Called 'Layer 3 Lite', it supports static IPv4 and IPv6 routing, inter-VLAN local routing and ARP (address resolution protocol) but not the dynamic routing found in more costly full Layer 3 switches.

Installation is as plug and play as it gets and you have two management options. The free SmartControlCenter utility discovers and manages multiple switches from one interface and provides tools to upgrade their firmware, with the ability to upload or download configuration files plus a quick access link to their web interfaces.

The switch's web console is well-designed and provides easy access to all features. It opens with a hardware overview showing the status of the three quiet cooling fans while switching to the device view page reveals a graphic of the switch with coloured port icons for their connection, speed and PoE status.

An impressive range of PoE controls is provided with a global usage threshold used to send out SNMP traps if it is exceeded. The Advanced PoE section provides a wealth of information about powered devices and usage.

We attached a range of wireless APs and IP cameras and could see their detected device class along with individual consumption in watts, milliamps and volts. Multiple timer schedules can be used to manage power to each port. These specify a start and end time period for every day or a date and time range and will turn power on or off for the ports they have been applied to.

Power controls let you decide which ports are allowed to supply power and you can set your own power limits on selected ports. One of four power priorities can be assigned to each port so if the drain reaches the switch's threshold, the ports with the lowest priority will be automatically switched off first.

The GS752TPP offers a wealth of standard switching features. Port, MAC and protocol-based VLANs are supported, the switch can create VLANs specifically for VoIP, and the Auto-VoIP feature identifies voice traffic using SIP or the OUI (organizationally unique identifier) of IP phones and automatically prioritises it.

QoS (quality of service) controls allow you to assign one of eight priorities to each port with a minimum bandwidth applied to each one. The switch supports 802.1p and can detect these fields and map the packets to the required priority queue allowing class of service (CoS) controls to be applied to similar types of traffic.

The switch offers great security as admin access can be controlled using RADIUS and TACACS+, 802.1x port authentication is supported and unauthenticated users can be passed to a guest VLAN with limited network access. It also provides DHCP snooping which blocks rogue DHCP servers and only allows clients to receive their IP addresses from authorized servers.

The GS752TPP beats the competition hands-down for value - Cisco's 48-port SG300-52MP, for example, offers Layer 2 and static routing features plus a 740W power budget but is twice the price. SMEs planning on a high-density PoE deployment that want great features and security won't find a more affordable alternative.

Verdict

The Netgear ProSafe GS752TPP beats the competition hands-down for value. SMEs planning on a high-density PoE deployment that want great features and security won’t find a more affordable alternative.

1U rack chassis;

700MHz MIPS-34Kc CPU;

128MB DDR;

32MB Flash;

48 x copper Gigabit with PoE/PoE+;

4 x Gigabit SFP;

760W power budget;

802.3af/at PoE;

104Gbps backplane capacity;

1.5MB packet buffer;

16K MAC addresses;

256 VLANs;

Internal PSU;

Netgear SmartControlCentre utility;

Web browser management;

Limited lifetime warranty

Usually, when we review a product, we're looking for the latest, greatest devices that do something that was hitherto unavailable.

In the case of an unmanaged switch, however, so little has changed in terms of the underlying technology that it doesn't age. The clue is in the name - unmanaged - there's nothing to configure and nothing to go wrong; in theory, it should just work.

And so, after seven years, the TL-SG1016D has hit our desk once again, with nothing changed save for sporting a newer TP-Link logo. But at under 50 excluding VAT, it's now a significant bargain, at less than half the price of equivalent models from other brands.

The key spec is that all 16 ports are gigabit (10/100/1000). With many other models, particularly then but to a lesser extent now, a couple of ports will give full speed, while the others languish at a more modest 10/100.

For the average user, if deployed correctly, it won't make a huge difference, but if the rest of your network can take advantage of the extra throughput, you'll notice a significant difference to the speed between, say a user's computer and a NAS device, which can often be a severe bottleneck.

All ports also offer MDI/MDIX negotiation. This means that whether you have CAT 5, CAT 5e or CAT 6 ethernet cables, the switch will work out the correct speed and traffic flow for you. Of course, again, your choice of cable will affect the performance of the network, but unless you've got some very creaky old CAT 5 cables in your network, you will feel the benefit.

As well as adjusting the speed, the switch can also correct the power throughput, meaning that the amount being consumed is significantly less than if all 16 were going hell for leather. This would matter more if Power over Ethernet (PoE) were supported, which it isn't, but it still matters. The whole system runs cool enough to be fanless anyway, which offers an additional saving in terms of cost to run.

Traffic management is offered with IEEE 802.3x flow control for full duplex and back pressure for half duplex and a total 32Gbits/sec switching capacity and 10KB Jumbo Frame Store and Forward transferring. There's also MAC address self-learning and auto-aging for added security.

What you don't get is pretty simple. You can't control access to individual ports by MAC address or time, you can't limit throughput to manually control speeds and as we've discussed, you can't power anything as there's no PoE.However, you can choose to have everything done for you in the most efficient way possible, and either put it on a shelf, or rack-mount it with the enclosed kit.

The indicators on the front show what each port is doing according to activity, and with multiple lights to show whether it has negotiated a 10/100 or 1000 (gigabit) connection.For a system that clocks in at less than many 8-port switches, there's a lot to recommend the TL-SG1016D. For the small to medium business that wants something that 'just works', it's ideal. If your infrastructure is aging, this represents a very cost effective way of improving throughput on the network. And with the proliferation of stuff from the Internet of Things, home power users will find it ideal as a way of managing traffic on struggling home networks.

With Gigabit Ethernet still not close to reaching the end of its usefulness, there's little to suggest that this won't do what you need from it for many years to come, and like many of the best devices, do it in a way that allows you to forget it's even there.

Verdict

Whether as the nerve centre for a home overflowing with gadgets, or the hub of a small office, this set-and-forget switch offers excellent value, even after 7 years

Dell EMC is on a mission to bring open networking (ON) to all walks of life and its N1108P-ON switch delivers it to campus and SMB networks. The concept behind ON is to do away with vendor lock-in by giving businesses the choice of which OS and apps they want to run on their switches.

Switches with proprietary firmware leave businesses at the mercy of vendors, as they'll release upgrades and new features whenever the mood takes them. Dell EMC removes this reliance; its rapidly-growing family of ON-ready switches support a wide range of 3rd-party Linux OSes.

The N1100 series of Layer 2 switches comprises six members, all supporting ON in hardware and offering a choice of Gigabit port densities. They currently come preloaded with Dell EMC's DNOS v6 firmware but are ready to run other Linux-based OSes as they become available.

Dell EMC has already launched its Debian-based DNOS v10 for its data centre switches, allowing engineers to run open-source Linux apps on the switch such as load balancers and firewalls. A scaled-down version for the N1100 family is expected to follow, along with support for partners such as Cumulus Networks.

They can be installed easily, as the switch can run the open-source ONIE (open networking installer environment) which is already provided for the N1108P-ON. We've seen this in action on Dell EMC's S6000 data centre switch, where we used it to boot the switch, install Cumulus Linux over the network and run scripts to enable selected ports.

The N1108P-ON targets SMBs that want to push wired network connectivity to areas such as small offices or meeting rooms but don't need a high port density switch. Its eight standard RJ-45 ports all support 10/100/1000Mbits/sec speeds while alongside are pairs of RJ-45 and SFP fibre ports supporting Gigabit only.

Support for PoE/PoE+ is available on the first four multi-speed ports. The switch has a 75W power budget so it can power a wide range of wireless APs, IP cameras and VoIP handsets.

Installation is simple: we used its Type-B USB serial port to connect to a Windows laptop and access the switch's CLI using HyperTerminal. On first boot-up, the switch runs a quick setup wizard to secure access, create a management VLAN, enable web browser access and activate SNMP.

CLI fans can continue using this for switch management or move over to the intuitive web interface. DNOS v6 provides a traditional set of L2 switching services and a lot more besides.

It differs from the DNOS v9 used in Dell EMC's data centre switches as it focuses more on user connections and can automate many network management functions. For example, its Auto-VoIP feature recognises IP phones and creates class-of-service (CoS) policies to prioritise voice traffic.

Compared with other L2 switches at this price point, the N1108P-ON is positively bursting with features. Security is impressive as along with up to 512 port and protocol-based VLANs, it supports MAC and IP address access control lists (ACLs) and per-port tiered authentication using 802.1x and MAC addresses.

The captive portal extends security to visitors and guests as it presents them with a fully customisable web login page when they connect to a port. The switch's local database supports 128 user accounts but for more, you can link up with an external RADIUS server.

We tested PoE functions with Linksys wireless APs and D-Link IP cameras. The switch had no problems powering them and from the web interface, we could see it had correctly identified their device class and allowed us to assign different power priorities to each PoE port.

The N1108P-ON initially looks pricey but this 12-port Gigabit switch is offering a lot of features for your money. It's a great choice for SMBs that want to push wired networking and PoE services to smaller venues and have the option of running other Linux OSes on it in the future.

This review originally appeared in PC Pro issue 278

Verdict

The N1108P-ON initially looks pricey but this 12-port Gigabit switch is offering a lot of features for your money. It’s a great choice for SMBs that want to push wired networking and PoE services to smaller venues and have the option of running other Linux OSes on it in the future.

1U half-width rack chassis

8 x 10/100/1000 RJ-45 ports, 2 x Gigabit RJ-45, 2 x Gigabit SFP

PoE/PoE+ (Ports 1-4)

75W PoE budget

24Gbits/sec backplane capacity

1.5MB packet buffer

16K MAC addresses

512 VLANs

CLI and web browser management

Limited lifetime warranty

Netgear is a high flyer in the SMB network switch market, offering an unbeatable combination of features and value. Its new XS716E aims to cement that position, bringing fuss-free 10-Gigabit (10GbE) switching within reach of any small businesses. The proposition makes a lot of sense: with so many servers and NAS appliances now 10GbE-ready, it's an instant way to expand network bandwidth and remove bottlenecks to critical applications.

Positioned at the lower end of Netgear's Smart Managed switch portfolio, the XS716E has sixteen fixed 10GBase-T ports, plus a single SFP+ fibre port. The latter is dual personality, sharing its link to the backplane with the sixteenth copper port, so you can't use both at the same time.

Don't get this switch confused with the older 16-port XS716T, which has two SFP+ dual-personality ports. Costing around 110 more, this model offers more advanced features, including 802.1x authentication and Auto-VoIP - plus Netgear's "Layer 3 Lite" which supports IPv4 and IPv6 static routing, VLAN routing and ARP.

Installation is as plug and play as it gets, with two management models to choose from: you can use either the switch's own built-in web server, or Netgear's free ProSafe Plus configuration utility. This provides full access to all the same features as the web interface, and can also discover and manage multiple switches from one interface.

The web console worked fine with Microsoft Edge and provides easy access to all features. You don't get any glitzy switch graphics (as you do with the XS716T), but it's easy enough to check on switch and port status. The port list shows the negotiated connection speed - you'll want to leave auto-negotiation enabled, as the only manual options are to disable ports or set them to Fast Ethernet speeds. Loop detection can also be enabled, and you can decide whether or not to allow the switch to be accessed by the ProSafe Plus utility.

VLAN options extend to the port and 802.1q varieties, and in both cases, you can use basic or advanced configurations. Advanced port-based VLANs, for example, allow ports to be members of more than one VLAN. Ports can also be placed in up to eight link aggregation groups (LAGs) for high-speed, fault-tolerant trunks. However, these can only be static and not dynamic, as the switch doesn't support 802.3ad and LACP (link aggregation control protocol) groups.

QoS (quality of service) lets you assign one of eight priorities to each port, with a minimum bandwidth applied to each one. The switch also supports 802.1p and can detect these fields and assign the priority defined in the packet.

Alongside all these technical capabilities, the switch also features some clever power saving technology, with support for the 802.3az EEE (Energy Efficient Ethernet) standard. With this enabled, the switch automatically detects cable lengths and adjusts power accordingly for shorter runs; disconnected ports have their power slashed to a minimum, decreasing overall power consumption and heat output.

That's not to imply that the XS716E is by any means a power hog. We measured the unit on its own consuming a mere 19W, and adding 10GbE links to our server and storage arrays saw consumption rise by only 2W per connection. The cooling fans run very quietly too.

The monitoring tab provides basic information about port traffic, with tables of bytes sent and received for each one although, unlike the XS716T, the XS716E doesn't support SNMP or syslog servers. However, if you have a network analyser you can use the port mirroring feature to send traffic from all selected ports to the one it's connected to for further analysis. There's also a cable tester that could prove useful: it'll check whether cables are functioning correctly, and if it finds an electrical fault and it'll tell you roughly where the fault is in metres.

The Netgear XS716E brings high-speed networking to the masses at a record-breaking price of less than 60 per 10GbE port - a value proposition other vendors can't hope to compete with. You don't get the full set of features found on more expensive switches, but for SMBs seeking a fuss-free speed boost for their network it's the perfect choice.

This review originally appeared in PC Pro issue 270

Verdict

The Netgear XS716E brings high-speed networking to the masses at a record-breaking price of less than £60 per 10GbE port – a value proposition other vendors can’t hope to compete with. You don’t get the full set of features found on more expensive switches, but for SMBs seeking a fuss-free speed boost for their network it’s the perfect choice.

1U rack chassis;

16 x 10GBase-T;

1 x SFP+ shared;

320Gbits/sec backplane capacity;

2MB packet buffer;

16K MAC addresses;

128 VLANs;

Internal PSU;

Netgear ProSafe Plus utility;

Web browser management;

Limited lifetime warranty

Data centres, HPC and big cloud builders they all have a voracious appetite for network bandwidth which will soon make the move to 100-Gigabit (100GbE) a necessity. Dell's Z9100-ON is the first 1U multi-rate 100GbE switch to market and it aims to tempt big businesses with a low price per 100GbE port.

The Z9100-ON uses Broadcom's Tomahawk ASIC and crams 32 100GbE QSFP28 ports into its 1U chassis. These can be used at their maximum speed for switch-to-switch or server links and Dell also offers a big choice of break-out cables.

The cables allow the switch to support up to 64 50GbE, 128 25GbE or 128 10GbE links. Dell also has a range of QSFP+ transceivers for converting ports to 40GbE speeds.

The ON' in the model name indicates the switch is a member of Dell's Open Networking switch family. As with the S6000-ON (web ID:25369), it supports the ONIE (Open Network Install Environment) so, along with Dell's DNOS, it can run a range of qualified Linux-based OSes.

Switch hardware

The Z9100-ON targets core data centre and top-of-rack (ToR) applications and has a backplane capacity of 6.4Tbits/sec. The price we've shown includes the DNOS software with all L2/L3 features enabled.

Along with the 32 100GbE ports, it also has a couple of 10GbE SFP+ ports to one side. Dell has thoughtfully provided these to allow you to connect legacy servers or switches without wasting a 100GbE port.

Along with 32 100GbE ports, the Z9100-ON also has a pair of 10GbE ports for legacy equipment

Hardware redundancy is good as the Z9100-ON has dual hot-plug PSUs and five hot-plug fan modules. All are offered with forward or reverse flows allowing the switch to support data centres with hot-aisle or cold-aisle cooling.

The switch's power supplies and fan modules support hot-aisle or cold-aisle data centres

DNOS 9 features

Dell's DNOS software provides all traditional switching and routing capabilities and there are no hidden costs. It includes advanced routing features such BGP routing and OSPF as standard and not as expensive optional extras.

The switch supports up to 128 link aggregation groups (LAGs) each comprising up to eight ports. This is great for fast switch-to-switch links and we tested this by easily connecting two Z9100-ON switches together over six aggregated 100GbE links.

The Z9100-ON support's Dell's VLTs (virtual link trunks) allowing two switches to be placed in a high availability domain. The new mVLT (multi-VLT) feature now allows LAGs to be used to create multiple redundant connections between logical VLT domains.

VLTs and mVLTs are the building blocks of Dell's Active Fabric which employs a spine and leaf architecture. It allows businesses to move away from expensive chassis based switches and expand capacity simply by adding new switches as spine or leaf nodes.

Performance tests

QLogic supplied its new FastLinQ QL45212 dual-port 25GbE HBAs which we fitted in two PowerEdge R730 rack servers. Using Windows Server 2012 R2 Teaming, we created 50GbE LACP links for each one and connected them to dedicated Z9100-ON switches joined over a 600Gbps LAG.

From the DNOS CLI, we created a high-speed 600Gbps switch-to-switch LAG

For storage, the first server was equipped with a 1.2TB Fusion-IO MLC SSD and three Samsung SSDs in a RAID0 stripe. On the second server, we had three Samsung SSDs also in a stripe which contained our test folder with three large files totaling 230GB.

To generate a high throughput, we used a Windows script to simultaneously copy the test folder to the other server's ioDrive and Samsung SSD array. Using the Windows Task Manager, we watched network throughput peak at 36.5Gbps with DNOS reporting switch-to-switch LAG utilisation of only 6 percent.

We ran a further test using the iPerf utility loaded on both servers and running bidirectional network throughput tests. Using four parallel client streams on each iPerf instance, we recorded a stonking cumulative network throughput of over 44Gbps.

Using Windows Teaming, we created 50GbE server links and watched iPerf generate a massive 44Gbps throughput

Pricey optics

The Z9100-ON is available now along with the full range of break-out cables and optics. We expect most early adopters will keep 100GbE within the data centre as long-reach QSFP28 transceivers are expensive.

The SR4 850nm optical transceiver has a 100-metre reach and costs a shade over 2,000. The LR4-Lite 1310nm optic supports up to 2kms and costs 10,706 while the full LR4 optic increases the reach to 10kms but drains your wallet to the tune of 14,609.

Switch management features during our tests were limited to the CLI, but support for Dell's Active Fabric Manager (AFM) and OpenManage Network Manager (OMNM) utilities will be implemented soon. AFM provisions switches in a network fabric, while OMNM uses SNMP to provide views of the physical status of switches.

QLogic has also announced immediate availability of its FastLinQ QL45000 series of server adapters. Along with the QL45212 HBAs we used for testing, QLogic offers the single-port QL45211 25GbE and QL45611 100GbE HBAs.

QLogic offers a choice of 25GbE and 100GbE server adapters

Conclusions

The discussions are over as 100GbE is right here, right now. As the industry's first multi-rate 100GbE 1U switch, Dell's Z9100-ON sets a high standard for the competition to follow.

Support for a wide range of port speeds allows early adopters to move to 25GbE now and upgrade to 40GbE, 50GbE or 100GbE when the price is right. Dell has delivered on its early promises to reduce the price per 100GbE switch port and support for ONIE allows you to pick and choose the OS you want.

Verdict

A high 100GbE port density, 1U form factor, multi-rate support, competitive costs per port and a choice of OSes – Dell’s Z9100-ON is the most versatile switch on the market

Chassis: 1U rack

Ports: 32 x 100GbE QSFP28

Backplane: 6.4Tbps full duplex

Forwarding capacity: Up to 4,400Mpps

Power: 2 x 750W hot-plug PSUs

Cooling: 5 x hot-plug dual-fan standard or reverse flow modules

Other ports: 2 x 10GbE SFP+, mini-USB console

Software: Dell DNOS 9

Netgear is on a mission to put affordable 10GbE switching solutions into the hands of SMBs, and the latest ProSafe XS728T shows it means business. This slimline 1U chassis puts 28 10GbE ports on the table at a price no other vendor will get close to.

As the flagship of Netgear's ProSafe Smart Managed Switch family, the XS728T has 24 fixed 10GBaseT ports and augments them with four SFP+ fibre ports. There are no dual personalities here, as the SFP+ ports are dedicated and canbe used alongside all 24 copper ports.

It's designed for small or medium- sized businesses that want high- performance core switching duties, or larger businesses looking for an affordable aggregation solution. With many of the latest NAS devices now 10GbE enabled, it's also an ideal solution for linking up a fast storage network.

At its foundation, the XS728T is an L2 switch with basic L3 routing capabilities. Called "Layer 3 Lite", it supports IPv4 and IPv6 static routing, VLAN routing and ARP, but not router discovery using IRDP (only the older 12-port XS712T supports this).

Unlike the entry-level XS708E (pcpro.link/255netgear), this switch can be fully managed via a web browser. Netgear also includes its Smart Control Center discovery software, which provides quick access to the web console, alongwith IP address setup and firmware and configuration download facilities.

Smart Control Center ran fineon our Windows 10 desktop, but we found that the switch's web console doesn't currently support Microsoft's Edge browser. We could log in to the switch, but Edge wouldn't display any details or allow us to access the menu tabs.

We had no problems with Firefox, Chrome or Internet Explorer, though, in which we found the web consoleto be well designed and easy to use.It opens with an overview of the switch hardware, along with system temperatures and the status of the four surprisingly quiet cooling fans.

We could see the status of system memory and the CPU, view a graph of CPU usage and switch to a device view with coloured port icons showing their connection status and speed. The switch's cable tests tell you whether they're functioning correctly orhave an electrical fault, and it will give an estimate of the cable length for 10GbE links.

The web console furnished us with lists of switch and port statistics for switch monitoring, and it can sendlog data to a syslog server. For SNMP, we used Paessler's PRTG Network Monitor to keep an eye on the switch and port status plus traffic throughput on each port.

Ports can be placed in up to eight static or dynamic link aggregation groups (LAGs) for high-speed and fault-tolerant trunks. Layer 2 features are extensive, with support for port, MAC and protocol-based VLANs. The switch can create VLANs for VoIP traffic and the Auto-VoIP option identifies SIP, H.323 signal and SCCP traffic and automatically prioritises it.

The quality of service (QoS) controls allow you to assign one of eight priorities to each port, with a minimum bandwidth applied to each one. The switch supports 802.1p and can detect these fields and map the packets to the required priority queue.

Security is on the money: Radius and TACACS+ can be used to tighten up admin access. Port authentication using 802.1x is supported and unauthenticated users can be passed to a guest VLAN with limited network access.

The switch is easy on the power supply. We measured it drawing 75W, with the load increasing by around 4W per 10GbE connection. Green Ethernet technology is supported, meaning that the switch will reduce power on ports where short cables are detected and, if the Energy Efficient Ethernet (EEE) feature is enabled, it will reduce power consumption on ports with low utilisation.

As the first 28-port 10GbE switch on the market, the XS728T sets a high standard for those vendors that dare to follow in its footsteps. SMBs looking for a path into high-speed networking simply won't find better value anywhere else.

Verdict

As the first 28-port 10GbE switch on the market, the XS728T sets a high standard for those vendors that dare to follow in its footsteps. SMBs looking for a path into high-speed networking simply won’t find better value anywhere else.

1U rack chassis

24 x 10GBaseT

4 x dedicated 10GbE SFP+

USB 2

560Gbits/sec backplane capacity

512MBsystem memory

3MB packet buffer

16K MAC addresses

Internal PSU

Netgear Smart Control Center software

Web browser management

Limited lifetime warranty

A major bugbear SMBs and enterprises have always faced with their choice of network switches is vendor lock-in. Switches running proprietary firmware leave them at the mercy of the vendor who will release upgrades and new features as and when they please.

You can choose the OS you want when you buy a server and Dell thinks you should able to do the same with your switches. The company is one of the few manufacturers to have designed its open networking (ON) switches to run a choice of 3rd-party OSes.

The family comprises four members and we review Dell's flagship S6000-ON preinstalled with Cumulus Linux. A key advantage of this partnership is that Cumulus Linux is built from the ground up as a data-centre grade networking OS and is capable of running virtually any Linux application you want.

Other OS choices include BigSwitch Networks Switch Light OS. This turns the S6000-ON into a true software defined network (SDN) physical layer switch managed by a separate SDN controller.

Ansible templates and playbooks can be used to automate the deployment of multiple switches

Switch hardware

The S6000-ON is ideally suited to high-performance, top of rack switching duties. It has 32 40GbE QSFP+ ports and has a claimed backplane switching capacity of 2.56Tbits/sec.

If you want more 10GbE, you can use the QSFP+ ports with breakout cables to split them out to four SFP+ 10GbE ports. This gives you up to 96 10GbE ports and still leaves eight 40GbE ports available.

Hardware redundancy is good as the S6000-ON has two hot-plug 460W power supplies. Cooling is handled by three hot-plug fans modules designed to work with the hot-aisle/cold-aisle cooling flows used in the latest data centres.

Ansible Tower provides a slick console for managing job templates and automation tasks

What Cumulus can do

The traditionally concept of an SDN calls for three layers with the application layer at the top and the physical network at the bottom. In a standard SDN, the central control plane is removed from the hardware and run in software as an SDN controller but with the S6000-ON, this remains on the switch and is managed by the Cumulus software.

Selected at the point of sale, Dell delivers the switch with the software preinstalled and with standard L2 switching and L3 routing enabled. Cumulus also offers a wealth of switch management and monitoring features.

It provides its own SNMP daemon allowing the switch to be monitored by any standard SNMP software. It can provide typical operational values such as processor, memory, port utilisation and traffic statistics. Cumulus also has an sFlow agent.

Cumulus supports MLAG (multi-chassis link aggregation) where both switches are active and if one goes down, the other assumes all primary functions and carries on with no service interruptions.

We could keep a close eye on the status of scheduled Ansible switch configuration tasks

What Cumulus can't do

If you're planning on using the switch to manage large storage networks then Dell's FTOS is a better choice of OS. Its iSCSI Optimisation feature allows the switch to recognise Dell Compellent and EqualLogic storage arrays and auto-configure flow control, Jumbo frames and CoS prioritisation.

FTOS offers a more traditional set of switching and routing services and also supports Dell's VLTs (virtual link trunks). Provided in Dell's standard S-Series switches, VLT is a more resilient and manageable alternative to stacking.

Cumulus Linux doesn't have a GUI front end so all management and configuration is done using the CLI (command line interface). However, we found the user manual very informative with typical Cumulus commands easy enough to understand and use.

Running Prescriptive Topology Manager on the switch allows it to check and verify physical cabling links

Deployment automation

With Cumulus on the case there's no need to configure and provision switches individually as this is automated using scripts. The switch has the Nano, vi and Pico text editors installed but, as this is Linux, you can load whichever text editor you like.

Deployment can be fully automated as when the switch first boots up, Cumulus knows it's new and can check URLs passed to it by a DHCP server. You can use the open-source ONIE (open networking install environment) to boot the switch, update it with the latest OS version and run scripts to enable selected ports, apply a license and email support staff to tell them the switch is ready for action.

For zero-touch provisioning, Cumulus can work with the Ansible automation tool and its scripts and templates. You're not restricted to Ansible as you can use whatever automation tool you want such as Puppet and Chef and although these require an agent on each switch. Cumulus has this covered too as they're installed by default and just need enabling.

Along with automation tools, the switch can run a wide range of other apps including DHCP and DNS servers or web servers such as Apache. However, bear in mind that some of these can be processor intensive so don't get carried away if your main priority is switching performance.

Conclusions

By separating hardware from software, Dell's open networking switches aim to disrupt the establishment and offer a versatile alternative. The S6000-ON and Cumulus Linux simplifies network deployment and automation tasks immensely, provides SDN future-proofing and delivers a wealth of 40GbE ports at a competitive price.

Verdict

Packed with 40GbE ports, Dell’s S6000-ON with Cumulus Linux is an innovative solution for data centres that tire of vendor lock-in and want more automation from their switches

Chassis: 1U rack

Ports: 32 x 40GbE QSFP+

Backplane: 2.56Tbps full duplex

Forwarding capacity: 1,462 Mbps

Power: 2 x 460W hot-plug power supplies

Cooling: 3 x dual-fan standard or reverse flow modules

Software: Cumulus Linux pre-installed

Dell has extended its portfolio of networking equipment with the launch of its new Dell Networking C9010 Network Director switch and companion C1048P Rapid Access Node. It also unveiled its Dell Networking S6100-ON in-rack datacentre switch.

The former two devices are aimed at mid- to large-scale campus environments. It claimedthat both of these would help in improving scalability and simplify management.

It added these would provide a single point of control for quality of service, policy provisioning and software upgrades, as well as for programming software-defined attributes.

The new Dell Networking C9010 Network Director is a next-generation multi-rate capable modular switch and is the first platform based on the new Dell campus network architecture. It scales up to 4,000 virtual ports managed centrally, with support for 10Gb and 40Gb Ethernet today, and a tool-free upgrade to 100GbE in the future to meet the growing bandwidth requirements, according to the firm.

Dell said the C9010 can be deployed as a traditional campus switching platform without the Dell Networking C1048P Rapid Access Node, serving to aggregate legacy switching platforms in wiring closets.

It added that when this was paired with the C1048P Rapid Access Node, this would extend the capabilities of the C9010 Network Director by providing enhanced power over Ethernet as well as user access and connectivity back to the C9010. The C1048P can be deployed either stand-alone or in a stacked configuration.

The Dell Networking S6100-ON is aimed at datacentres and combines multi-rate connectivity with chassis-level modularity and open networking. Through multiple expansion modules, customers have the flexibility to deliver various combinations of port speeds ranging from 10 100 GbE with both QSFP28 and CXP port choices for flexibility in deployment.

The Dell Networking C9010 and C1048P switch and node will be available in October, 2015 with the Dell Networking S6100-ON switch available in the first quarter of 2016.

ARM has launched a free operating system aimed at powering Internet of Things (IoT) devices, as well as tools to help developers create them.

The software is offered for free in a bid to boost uptake of IoT products.

The mbed OS is an open source operating system designed to run on ARM's 32-bit Cortex-M and contains security, communication and device management features to enable the development of production-grade, energy-efficient IoT devices, according to the firm.

The OS will also support a number of wireless standards, such as Bluetooth Smart, 2G, 3G, LTE and CDMA cellular technologies, Thread, Wi-Fi, and 802.15.4/6LoWPAN along with TLS/DTLS, CoAP, HTTP, MQTT and Lightweight M2M.

The OS forms part of ARM's mbed IoT Device Platform and also features the mbed Device Server that required server-side technologies to connect and manage devices in a secure way. The firm said this would also provide a bridge between the protocols designed for use on IoT devices and the APIs that are used by web developers.

It claimed this would simplify the integration of IoT devices into cloud frameworks that deploy big data analytics on the aggregated information.

"ARM mbed will make this easier by offering the necessary building blocks to enable our expanding set of ecosystem partners to focus on the problems they need to solve to differentiate their products, instead of common infrastructure technologies. This will accelerate the growth and adoption of the IoT in all sectors of the global economy," said ARM chief executive Simon Segars.

The mbed IoT Device Platform is free for developers and for deployment, supported by an ecosystem of more than 100 OEMs and major chip vendors.

Launch partners for the mbed IoT Device Platform include Atmel, CSR, Ericsson, Farnell, Freescale, IBM, KDDI, Marvell, MegaChips, MultiTech, Nordic Semiconductor, NXP, Renesas, SeeControl, Semtech, Silicon Labs, Stream Technologies, ST, Telenor Connexion, Telefonica, Thundersoft, u-blox, wot.io and Zebra.

The platform will be available to mbed partners in the last quarter of this year for early development, with the first production devices due in 2015.

Intel has agreed to buy Avago Technologies' Axxia networking business for $650 million.

The all-cash deal, which has been approved by the boards of directors for both Avago and Intel, is expected to close in the fourth quarter of this year, subject to regulatory approvals.

The unit, which made 113 million in revenues last year and employs 650 people, will help Intel in its bid to provide processors to service providers building more intelligent networks. The firm makes systems-on-a-chip (SOCs) for networking infrastructure.

"With this acquisition, Intel gains a compelling product line and talented team to continue to deliver on this long-term vision," said Rose Schooler, vice president of Intel's data centre group, in a blog post.

She added that the opportunity was "significant and wireless access is a critical component of the mobile network, which is part of a growing network market segment estimated to represent a $16 billion silicon market segment".

Intel already supplies the vast majority of processors in PCs and servers and is looking now to bolster its share in markets such as networking, the Internet of Things and mobile devices. With the latter, Intel has been caught on the hop by the likes of ARM. Such a move would help Intel decrease its reliance on an uncertain x86 market.

"In the same way that Intel helped to transform the data centre into a business enabler, Intel can help transform networking by providing new technologies and a portfolio of solutions that allow for new innovative services and efficient scaling of a more flexible, cost effective network infrastructure," said Schooler.

Axxia was part of LSI, which Avago acquired for $6.6 billion in May. That month, Seagate signed a definitive agreement with Avago to buy LSI's flash storage business for $450 million.

The Chinese telecoms firms at the centre of a political storm, ZTE and Huawei, have hit back at allegations that their equipment could be used to spy on American companies and government agencies.

In a statement, ZTE reaffirmed its commitment to work with the US House Permanent Select Committee on Intelligence, government agencies and the private sector to address all security concerns.

David Dai Shu, ZTE's director of global public affairs, said "ZTE appreciates the Committee's recognition that ZTE has offered US carriers a Trusted Delivery Model solution. ZTE will work with the Committee, US government agencies, and ZTE's US customers to identify and deploy the most effective equipment cyber-security measures possible. ZTE is committed to assuring US carriers and US government agencies its equipment is safe."

The Shenzhen-based company also intimates in its statement that it is unfair to single out ZTE and Huawei simply because they are the two largest Chinese companies supplying infrastructure to the US.

"It is noteworthy that, after a year-long investigation, the Committee rests its conclusions on a finding that ZTE may not be free of state influence.' This finding would apply to any company operating in China ... given the severity of the Committee's recommendations, ZTE recommends that the Committee's investigation be extended to include every company making equipment in China, including the Western vendors. That is the only way to truly protect US equipment and US national security," added Dai Shu.

ZTE's commitment follows news that networking specialists Cisco split with the company over allegations the Chinese firm had sold Cisco branded products to Iran. The US House of Representatives Intelligence Committee has also warned American companies and government agencies to avoid using ZTE and Huawei. The committee claimed in a report published on 8 October that the two firms represent a national security threat to the US. However in the UK, Huawei is also a major supplier to BT and Everything Everywhere.

Huawei has been even more forthright in its response to the committee's report. A spokesperson for the company said: "The United States is a country ruled by law, where all charges and allegations should be based on solid evidence and facts. The report conducted by the House Permanent Select Committee on Intelligence ... failed to provide clear information or evidence to substantiate the legitimacy of the Committee's concerns."

"The report released by the Committee today employs many rumors and speculations to prove non-existent accusations ... We have to suspect that the only purpose of such a report is to impede competition and obstruct Chinese ICT companies from entering the US market," the spokesperson added.

HP is to partner with F5 Networks to cut down the time involved in configuring and deploying virtualised server applications on enterprise infrastructure.

The new initiative sees F5's application management tools combining with HP Virtual Application Network (VAN) fabric. The platform offers a set of plugins and configuration tools which let administrators create policies and profiles for configuring virtual machines and integrating them into an organisation's infrastructure.

Using the two technologies should see application deployment times cut from months to minutes, according to estimates by HP.

"We have separated the control plane from the devices themselves," said Bethany Meyer, general manager of HP Networking at the Interop conference in Las Vegas. "It makes networks programmable."

Layering automatic network configuration on top of HP's managed network products should increase the performance of popular application servers, such as Microsoft Exchange 2010, VMware-based backup and disaster recovery, and application acceleration, according to HP.

According to HP Networking's director of marketing Kash Shaikh such a platform would remove the need for systems administrators to configure settings manually in virtual machine deployments. He added that using command line interfaces to set up network configurations can require up to quarter of a million entries per server and it was easy for human input to create errors.

"You can still have one error in a thousand and end up with over a thousand errors," he said. Shaikh added that this would be the "beginning of the end for the command line interface."

The integrated technology means that networks can automatically sense the application running on the network and adapt. Meyer said that video could be displayed properly by analysing network traffic, seeing video was being passed through and adjusting the network to optimise performance.

Such adjustments could decrease spam on networks running Exchange 2010 by up to 70 per cent and decrease the time needed to move a virtual machine from datacentre to datacentre from 20 minutes to around 38 seconds, according to Meyer.

The release of HP's Virtual Application Networks is earmarked for around June. HP would not disclose pricing for the new platform.

HP said it will release further information about the platform over the next few months. It also plans to use F5 technology to accelerate other enterprise applications other than Exchange and VMware.

Networking equipment maker Extreme Networks revealed its 10 Gigabit (10G) Ethernet copper modules for its BlackDiamond X8 core switch and new fixed copper versions of its Summit X670 top-of-rack switch. The new products are being pushed by the comapny as a primary way for datacentres to upgrade their networks.

The products were unveiled at the Interop conference in Las Vegas and the new high density copper switches are optimised to take advantage of servers based on the Intel Xeon Processor E5 Family and 10G LAN on motherboard (LOM) devices.

According to research company Dell'Oro Group, 10G copper switches, as a ratio of global ports shipped, will become the primary choice for cloud and datacentre network upgrades in the next two years.

Huy Nguyen, senior director of product management at Extreme Networks said that Investment protection is critical to new datacentre operators.

"Datacentre operators are always looking for ways to reuse cabling infrastructure, when upgrading their networks to 10G. Rip-and-replace is the last option for most datacentre operators," said Nguyen.

He added that the new BlackDiamond X8 blade and Summit X670 copper switch allow data centre customers to deploy "the most cost-effective 10G options using standard RJ45 copper jacks."

The new Summit X670 switch supports 48-ports of wire-speed 10G copper RJ45 with 40G uplink capabilities and the company said the product is expected to ship this year. The new BlackDiamond X8 copper blade supports 48-ports of wire-speed 10G copper RJ45 with 384 total ports inside the chassis. These products can be used by datacentre and cloud operators in modular and fixed configurations.

Cisco appears to have turned a corner after posting record revenue alongside solid profit growth in the company's second quarter results.

Net profits rose 43.5 per cent reaching $2.2 billion (1.3 billion) for the three months leading up to 28 January. Sales figures hit $11.5 billion.

The networking giant had a turbulent 2011, a year in which there were widespread job cuts and a major strategic refocus, although the company's UK CTO recently told IT Pro no more employees would be sent packing in 2012.

We are executing well on our three-year plan to drive earnings faster than revenue.

Products including Flip cameras and the umi home videoconferencing have been ditched along the way, after Cisco decided to turn back on its foray into the consumer sphere.

The company had repeatedly disappointed investors, but in the past few months Cisco has performed considerably better.

The results hinted at weaknesses in public sector buying power, largely thanks to austerity plans of Governments across the world, CEO John Chambers revealed in a conference call. Yet the company's core products, including switches and routers, brought in plenty of revenue.

Cisco's switching business saw eight per cent revenue growth. Switching gross margins have returned to where they were two years ago, again hinting the company is getting back to what it's best at. Routing also grew eight per cent in revenue.

"We are executing well on our three-year plan to drive earnings faster than revenue. Our operational focus continues to yield positive results - we hit our billion dollar expense reduction a quarter early - and our ongoing innovation enables our customers to solve their critical business needs," Chambers said.

"You will continue to see a focused and aggressive Cisco that is helping our customers use intelligent networks to transform their businesses."

The company is also making gains in the data centre, with its UCS servers acquiring 10,000 customers overall, with 3,000 in Europe.

Cisco has been benefitting from competitive weakness too, with Juniper Networks' fourth quarter results showing revenue below expectations and net income falling to $96.1 million from $190.2 million the year previous.

Read on for our Q&A with Cisco's vice president for enterprise Laurent Blanchard.

For anyone who's been watching Cisco's development over the past few years, they'll know the company went through a somewhat chaotic period in 2010 and 2011.

In a bid to galvanise investor confidence, the company distanced itself from the consumer sphere, streamlined operations and went back to focusing on the enterprise.

This translated into the death of Flip cameras and umi home videoconferencing, job losses and greater embracing of technologies tied to the network.

Three years ago, no one would have believed Cisco would play a role in the server market.

We caught up with Laurent Blanchard, the networking giant's vice president of enterprise, at this year's Cisco Live event to ask why IT departments should get excited about Cisco again.

Cisco is still fairly new to the server space, having only entered the market with UCS (Unified Computing and Servers) a few years back. What can you say to IT departments to convince them Cisco is a worthy contender in this industry?

I'm not here to convince you if you are not ready to be convinced but I'm going to give my point of view.

Three years ago, no one would have believed Cisco would play a role in the server market. Actually, what we did was apply our knowledge of driving a huge amount of data through something the network into a server. We redesigned a server from scratch. When you have a legacy of many years, it is very difficult to do this. But the structure of the UCS is new from the design perspective.

In two years, we have just signed the 10,000th UCS customer. Who would have though that we would do that three years ago?

We can differentiate ourselves and the market is actually accepting it very well.

What about bringing in more storage capabilities? Is that something Cisco will do in the future rather than just partnering?

When you look at what is happening, people are moving to the cloud and we've found that there are three levels of services for infrastructure that have to be provided: compute, storage and network.

We believe we have a very strong knowledge of servers, we have 25 years+ in the network and today we believe in storage we are better off partnering with very big companies such as EMC.

We work with NetApp and we work with others as well. We have building blocks together with NetApp in the same way we have with EMC. You can say one is larger than the other one but our approach in the cloud is to drive and to work in an ecosystem.

I think the industry is understanding that the previous way of looking at IT has changed. Now the three elements are extremely integrated and you need to have a huge amount of knowledge in intelligent networks, you have to understand very well how you can virtualise in the server and therefore have the storage as capacity for those two other elements.

It is normal that the industry is going in this direction. Three years ago nobody would have believed it. I think Dell is looking at it as well, like HP, like the others.

With companies coming together in clusters to provide services, what would you say to those worried about vendor lock-in, or siloed IT architectures?

This is not what people have on their minds. First of all, a CIO of a very large organisation, they have an environment with many providers, many suppliers. They may have HP servers and Brocade solution, they don't have only one provider.

The industry is understanding that the previous way of looking at IT has changed.

Our value is that we know how to get all these pieces together and working extremely well.

There was a lot of talk about innovation during the Cisco Live keynote. Can you convince IT departments that you really are innovating when so many smaller players are making big strides?

Our key differentiator is that we know how to manage huge volumes of data transfer in an intelligent manner, in a secure manner.

If you look at the cloud, in 2010 people were saying there is roughly $30 billion investment being made into the cloud. By 2014, it will be at $100 billion and 50 per cent of the compute consumed will be cloud based.

Now if you agree with this, and if you agree with the fact that the volume of data is going to explode, and if you agree the security issue is going to be important you need to have the right network. This is where you put the most intelligence.

The innovation is to continue investing into development of the network and to answer major trends like bring your own device (BYOD), the internet of things, smart grid. Smart grid is huge business. We apply our innovation into the energy market, the smart grid.

We can make video data intelligent. You can do a search on videos now, for example. So you can search for what someone said or for keywords.

Going back to cloud, how much traction have you got within service provider data centres?

We do work with the traditional service providers, like Amazon and Google.

A lot of the talk at Cisco Live has been around providing better than 'good enough' networks, rather than premium ones like Cisco claims to offer. But a lot of IT managers are happy with good enough as long as it has got a low barrier of entry. Do you worry some of the cheaper providers are going to steal business away from you in the current climate?

I think the trend from the IT and CIO/IT manager side, the big one, is the move from Capex to Opex. When you look at the cost of running an operation, most of the cost is not the initial investment, it's running it.

You can go and buy a very cheap server with no security built into it, but it is going to cost you a lot of money after that.

But then you have the business angle which is continuously pushing you to get a new environment, new devices to be supported. Actually, what you're looking at is elasticity and agility at a cost which is the lowest cost.

We're clearly looking at the total cost of ownership (TCO), the cost of running the operation, as well as doing no compromise on security.

You can go and buy a very cheap server with no security built into it, but it is going to cost you a lot of money after that. Security is an environment where we feel there is no compromise and today it is one of the top asks of the CIO.

Can you prove that your TCO is lower than others?

We have very large companies who've done this. Let's take banking firm ING, they moved into a solution using UCS for virtualisation and they've been consolidating many servers they have IBM, HP, Dell servers. They have been consolidating using Cisco technology UCS with a VMware hypervisor and EMC. The cost of what they are running is lower than what they had before.

I don't like overall studies covering this because each environment is different, but I can tell you we would not have gotten to 10,000 customers if the competition was cheaper.

I've been at HP where I sold x86 and you would sell it, sell it, sell it. But if you don't have any differentiators, you are under an additional pressure margin. This is not where we want to play.

Cisco is looking to help IT managers deal with ever-expanding traffic, today announcing upgrade options so customers can achieve 40 and 100 Gigabit Ethernet capacities in their switches.

The networking giant is offering 40GbE and 100GbE capabilities in its Nexus 7000 switches via two separate modules, alongside a 40GbE module for its Catalyst 6500 products, all delivered by a tonne of 10Gbps lanes.

The latter focuses more on campus setups, the former more on enterprise data centres.

If we got lazy we would get left behind.

The upgrade modules for the various switches will please those looking to cope with ever increasing bandwidth, or IT departments hoping to kick start cloud or big data projects.

Customers appeared at Cisco Live in London today to talk about why they were excited about the announcements.

"We're seeing higher and higher peaks of traffic, so the next natural step is to go to 40G. The announcement on that was really the right timing for us. It is time to look for higher capacity links as we're really going to need them," said Jeroen van Ingen, from the University of Twente's ICT service centre.

Malcolm Days, head of infrastructure services at Warwick University, said the educational body sees ever increasing volumes of data and so needs the extra capacity.

"When 7,000 students are watching video and its moving towards high definition, that's using up a lot of bandwidth," Days said. "As long as the modules are coming when we need the bandwidth, we're happy."

The Catalyst 6500 40GbE module will be release in April 2012, whilst the two Nexus 7000 modules will be released in the second quarter.

Innovator or slow starter?

Cisco was keen to point out that it was leading the pack by getting 40GbE and 100GbE to market, noting Juniper was not supplying such capabilities on its switching platforms.

"Let's compare apples to apples," said Ram Velaga, vice president of product management for switching.

Cisco is convinced it has come along at the right moment with its 40GbE and 100GbE modules, as now is the time when customers are looking to upgrade.

"Sometimes we are ahead of the competition, sometimes we are not," said Christian Korff, director of sales for Cisco's Borderless Networks division. "But if we got lazy we would get left behind."

Korff also claimed Juniper's main selling point was that it could offer a single operating system to manage networks Junos OS. However, Juniper customers will really have to deal with numerous operating systems, he claimed.

Force10, now owned by Dell, was the first to get on the 40GbE switching case way back in 2010. At the time, it distanced itself from 100GbE, saying most customers wouldn't want to shell out to support the premium standard.

Yesterday at Cisco Live, the networking vendor claimed WAN optimisation was just a feature, hinting it may not be a market at all in the coming years.

The London Underground is set to have Wi-Fi in place across 120 stations in time for the 2012 Olympics, despite a delayed tender process.

Transport for London (TfL) confirmed it would announce the chosen service provider in early spring, even though it initially planned to have chosen a bidder by the end of 2011.

"We are in the final stages of the tender process," said Gareth Powell, director of strategy and service development.

"London Underground is continuing with preparations to install the necessary infrastructure and is on schedule to complete the project as planned."

In March last year, TfL called on service providers to bid for the contract following a successful trial at Charing Cross station BT Openzone that has been running since November 2010.

In February 2011, it was rumoured Chinese tech giant Huawei was in talks to provide a mobile network on the Underground, but nothing has yet materialised from the reports.

Olympics organisers have sought to guarantee technological infrastructure is ready for the big event this summer. Ofcom revealed yesterday that demand for spectrum was going to double during the Games.

The regulator explained how public sector organisations would be sharing their spectrum to help deal with demand.

Mobile coverage will not be coming to the Underground in time for the Olympics, however.

TfL hoped mobile companies would be able to fund the initiative, but 3, Everything Everywhere, O2 and Vodafone said they had given up on the plans as they were unable to agree on a viable proposal.

Cisco will not be reviewing its pricing strategy in 2012, even in the face of reduced IT budgets across the globe.

Customers IT Pro and others have spoken to have cited Cisco's premium prices as off-putting, when compared to cheaper offers from competitors.

Cisco, which is in the midst of major strategic changes, will not budge on pricing anytime soon though, according to Ian Foddering, chief technical officer and technical director for Cisco UK.

"We will always look to remain competitive but there will not be any wholesale change," Foddering said.

Cisco's refusal to budge on pricing will disappoint IT departments looking to make more out of less. UK councils, for instance, look set to have budgets cut by as much as nine per cent in 2012, according to a Socitm report released yesterday.

We will always look to remain competitive but there will not be any wholesale change.

CEO John Chambers called on Cisco to refocus on it's core strengths namely networking - in 2011 after repeatedly disappointing investors.

This has led to various consumer products being sent to the scrapheap, including Flip cameras and the umi videoconferencing offering.

Yet a pricing rethink does not feature in Cisco's re-organisation plans.

And now for some good news...

Cisco's refocusing has led to job losses too, but Foddering told IT Pro he does not believe there will be any more redundancies this year, even in the face of a difficult financial climate.

"I don't foresee any further changes," he added.

"I'm bringing on board six new apprentices and five new interns actually the big challenge right now is finding the right people."

Whilst he could not provide any figures for how many UK workers were made redundant, Cisco spokespeople said the majority of job losses were in the US.

Many people in the UK who did lose their roles did so voluntarily, according to the Cisco spokesperson.

The company will hold its premier event, Cisco Live, in London at the end of the month.

Check back on IT Pro for all the latest news from the event when it kicks off on 31 January.

The switch to IPv6 from IPv4 is gathering momentum with the announcement that the World IPv6 Launch day will take place on 6 June.

A host of tech giants have jumped on board, saying they will ensure their technology is compatible with IPv6 traffic, organisers the Internet Society said.

Companies including Cisco, Facebook, Google and Microsoft are taking part, as IPv4 addresses get close to running out.

As there are more IPv6 services, it becomes increasingly important for companies to accelerate their own deployment plans.

Outside of compatibility problems associated with this, cyber criminals could use IPv6 traffic to bypass firewalls and security appliances, some of which will not have been configured to pick up on IPv6.

"The fact that leading companies across several industries are making significant commitments to participate in World IPv6 Launch is yet another indication that IPv6 is no longer a lab experiment; it's here and is an important next step in the internet's evolution," said Leslie Daigle, the Internet Society's chief internet technology officer.

"And, as there are more IPv6 services, it becomes increasingly important for companies to accelerate their own deployment plans."

Last year, tech companies organised World IPv6 Day, which saw them testing out technologies in preparation for the switchover.

All businesses were called on to use the day as a test for the shift.

"Last year's industry-wide test of IPv6 successfully showed that the global adoption of IPv6 is the best way to keep web devices communicating in the future," said Jay Parikh, vice president of infrastructure engineering at Facebook.

"Permanently enabling IPv6 is vital to keeping the internet open and ensuring people stay connected online as the number of web users and devices continues to grow."

At Google, almost all its services will be available over IPv6 permanently from 6 June for everyone.

"It will take years for the internet to transition fully to IPv6," said Erik Kline, a Google IPv6 software engineer in Tokyo, in a blog post.

"Permanently enabling global IPv6 access to Google services has been our goal since we first began our IPv6 project more than four years ago."

Parliament is looking to get hooked up with fresh Wi-Fi and networking equipment, according to a tender document published online.

Parliamentary ICT is looking for a wide-scale refresh for both the House of Commons and the House of Lords, including a telephony system replacement to cover around 6,000 ports.

The same provider who installs the telephony technology may also be tasked with bringing future services, including unified communications and convergence with other parliamentary networks.

The supplier will be expected to demonstrate value for money, plus the capacity for flexibility.

The tender document also revealed Parliament is on the hunt for a managed Wi-FI network for members of both houses as well as for visitors.

"The new network infrastructure must be resilient and have sufficient capacity to carry additional network traffic at a later date," the document read.

"This will allow Parliament to achieve its long term vision of network convergence and delivering the strategic aim of enabling 'to connect members, the public and the administration to the information and services they need from anywhere, at any time and from any device; to reduce the cost of ICT; and to provide new opportunities and pathways for greater efficiency and effectiveness for Parliament.'"

One provider may be able to secure both 'lots' for the telephony and Wi-Fi deployments, but Parliament is open to using two vendors for the separate projects.

"The supplier will be expected to demonstrate value for money, plus the capacity for flexibility, innovation and continuous improvement throughout the life of the contract, together with the ability to manage the planned/unplanned small and large scale changes," the tender document read.

The tender revealed Parliament expects implementation to kick off in September 2012 and be completed by March 2014.

Major vendors like BT and Virgin will most likely be interested in bidding, although major mobile operators are capable of bringing the goods.

O2 yesterday revealed it had won a contract with Network Rail to refresh the company's current Cisco switch and router estate across 450 sites.

Suitors will have to bid to a Government that has become tougher on suppliers since the days of Blair and Brown's Labour rule, which was criticised for wasting money on IT.

BT has secured a lucrative deal with the European Parliament, which will see the telco giant provide networking under two framework contracts.

Network equipment and applications will be delivered to the EU body, alongside professional services for those products across three main sites in Brussels, Luxembourg and Strasbourg.

BT will provide the technology to support a host of telecoms services, including VoIP, unified communications, videoconferencing and telepresence.

BT put together a competitive and strong offering which really stood out.

The deal has been valued at 120 million (100 million).

"BT put together a competitive and strong offering which really stood out," a representative for the European Parliament said.

"Their highly-skilled integration experts and a vendor agnostic approach should enable us to implement all communication infrastructures and services according to our Call for Tender specifications. This makes me confident that we have chosen the right partner."

The deal marks another high point for BT's Global Services division, which is responsible for a sizeable chunk of the telecoms firm's revenue.

In early 2009, the division was actually holding back BT revenue. At one stage it was reporting losses of 124 million. In 2008, BT said it would cut 10,000 jobs due to Global Services' poor performance.

Fast forward to the company results in November 2011, profits had risen 15 per cent and the division was making money thanks to some big deals.

BT has scored a number of contracts in the public sector this year, including one to become a provider for the Public Sector Network.

"Our extensive portfolio of services and our professional services capabilities were instrumental in winning these deals," Luis Alvarez, president for EMEA and Latin America at BT Global Services, said today of the European Parliament deal.

"Our consultancy and systems integration expertise is increasingly recognised in the industry. I want to thank the European Parliament for the trust they put into BT."

CASE STUDY Real Madrid have been known to chuck inordinate sums of money at twinkle-toed footballers. In recent years, spending sprees haven't brought back the glory days. Even with Jose Mourinho at the helm, the club is still playing second fiddle to Barcelona.

At least Los Merengues should have a more high-tech stadium though. Thanks to a deal with Cisco, even if the likes of Ronaldo and Kaka aren't dominating on the pitch, fans will be appeased with the exciting mobile experiences they will be able to enjoy at the Bernabeu.

The club's IT team and the networking giant have been working on implementing some Cisco Connected Sports Solutions to give fans more to do once in the stadium other than watching the game. End users will be able to log-on and access apps for interacting with the stadium, whilst HD video will be on display in the stadium, meaning you'll be able to spot what interesting new hairdos players are sporting in really fine detail.

This is one IT project with a few missing pieces right now.

The Connected Stadium Wi-Fi network should provide improved non-data services too. Cisco thinks that by off-loading all data traffic through the cellular network, bandwidth will be freed up for phone calls and texts.

A trip to the Bernabau should soon be a more immersive experience. But 'should' and 'soon' are the operative words here. This is one IT project with a few missing pieces right now.

No end in sight?

Despite the promise of a fantastic IT project, all is not as rosy as it seems at the Bernabeu. Planning hasn't been perfect. For starters, even though Cisco was keen to push out a release this week, the initiative isn't complete yet and there is no confirmed end-date.

"Cisco and Real Madrid are working together to define the optimal implementation schedule, one that takes into consideration all planned events at the stadium," said Stuart Hamilton, director for business development in sports and entertainment at Cisco. "Unfortunately, we do not have an exact date at this stage."

Real Madrid hasn't figured out how it is going to market the service either. It appears the planning phase didn't look at the whole picture.

"[The] business model is now being defined and we cannot exclude any scenario. The range of possibilities is huge and we should be accurate in the proposal we launch to spectators," said Real Madrid CIO Enrique Uriel.

"What is really sure is that either spectators or visitors will need to log-in in an application that will launch all the services."

What services though? There's been a lack of detail on what the Cisco set-up is actually supporting. Whilst there will be ways to interact with the stadium, neither Real Madrid or Cisco have been forthcoming in what format this will be delivered. They say apps, but there has been no mention of what these apps will do.

Of course, Cisco simply provides the networking back-end for the entire solution. It seems it's Real Madrid that hasn't quite figured out what the whole shebang will look like. When asked what other vendors had supplied infrastructure for the project, Hamilton said that was "still to be determined."

Uriel said the project was "the result of an actual study of the scenario and all details have been considered in the design phase." Yet the admittances that certain decisions have not been made indicate the entire project, from its design to its implementation, was not outlined from the start.

The networking side has been dealt with, it's just other pieces of the jigsaw are lacking. Do the Real Madrid faithful care about the quality of the network though? No. They just want to tuck into the potentially exciting services that the overall deployment will offer.

Despite the initiative's current shortcomings, at least the network will be up to scratch. Cisco said this was more than just "super dooper Wi-Fi" it was bringing to the Bernabeu.

"What Cisco has done here is used the set of tools that we have (Rogue AP detection, Radio Resource Manager, Clean Air for RF interference detection, Band Steering to move clients to the appropriate radio band) along with new innovations such as narrow beam High Gain Antennas and new RF tuning techniques to optimise the Wi-Fi network for the most demanding environments," Hamilton said.

Grand ambitions need substance too.

"The connection speeds of the client will depend on the type of client and what it supports, but the Connected Stadium Wi-Fi network supports the 2.4GHz and 5GHz bands with 802.11g and 802.11 a respectively as well as 802.11n on both bands.

"For example, speedtest measurements using the speedtest.net application on iPhones have measured upload and download speeds in the 15-20Mpbs range in some stadiums."

If those speeds prove to be accurate in the real world, that's some speedy mobile internet. Better than what a lot of people get in their own homes.

The actual workforce to manage the task has been established too. To nail an IT project for an entire stadium, especially one as massive as the Bernabeu, you need quality planning and the right people on board. In this case, given the range of different technologies being used, particular skills were required.

"The team selected to deploy the project by both Cisco and Real Madrid are people with deep knowledge and skilled in all the domains (Wi-Fi, networking, TV, cabling, mobile applications, business, etc) that compose such a projects like this."

It's clear some headway has been made to achieving the end goal. However, just as with Mourinho's men, the package looks promising but the team is yet to deliver. Grand ambitions need substance too.

As any manager will tell you, whether in IT or football, having a complete strategy in place from the start is a big boon. If Real Madrid can fill in the gaps, however, the Bernabeu may well become what its owners hope it will be: one of the most technologically advanced stadiums in the world.

Cisco has forecast revenue and earnings above Wall Street expectations as it looks to bounce back after some tough times.

Analysts had expected conservative quarterly guidance, given the economic uncertainty.

Chief executive John Chambers said that budgets of large customers as well as governments were better than expected.

The world's biggest networking equipment maker projected a seven to eight per cent rise in fiscal second-quarter sales, translating to $11.13 billion (7 billion) to $11.2 billion in revenue - matching or slightly ahead of the $11.14 billion expected, on average.

Excluding items, Cisco predicted earnings per share of 42 to 44 cents in the quarter, beating the average forecast of 42 cents, signalling its months-long turnaround was bearing fruit.

But Chambers, who kicked off a months-long overhaul of the company to save $1 billion through layoffs and asset sales, warned that global uncertainty persists and it remains tough to predict market conditions.

"There will always be challenges," Chambers said. "We are watching very closely the developments in Europe and the global economy, public sector spending, India business, and the fallout from the flooding in Thailand."

Cisco competes with Juniper Networks, Brocade Communications, Alcatel-Lucent and Huawei, many of which have grabbed market share from the erstwhile high-growth Silicon Valley darling and chipped away at its margins.

Chambers vowed to make life difficult for rivals, in particular China-based Huawei. "In the past we have been a little too gentle," he said.

For the time being, unlike most of the competition, Cisco looks to be on track to revive growth and return to its old glory after it slashed its long-term targets and laid off thousands of employees.

Juniper Networks forecast disappointing fourth-quarter results, while Alcatel-Lucent scaled back its profitability goal for the year as telecom operators hold back spending in response to mounting economic uncertainty.

"The key takeaway is that Cisco executed well in a tough environment," Edward Jones analyst Bill Kreher said.

"Overall, the print looks clean with margin and operating profit upside relative to our estimates that are above the Street," Brian White of Ticonderoga Securities said.

Cisco beat its own margin expectations in the first quarter with non-GAAP gross margins at 62.4 per cent, above its target of 61 to 61.5 per cent.

"Gross margins appear to be stabilising. We view that as an important step in the Cisco turnaround story," Kreher said.

Cisco's shares extended gains after Chambers' comments, rising four per cent to $18.30 in extended trade, after closing down 3.8 per cent.

On Wednesday, Cisco reported quarterly earnings per share that beat estimates, signalling that efforts to revive growth are beginning to pay off.

The company reported adjusted earnings of 43 cents per share for the fiscal first quarter ended 29 October, compared with the average analyst forecast of 39 cents, according to Thomson Reuters I/B/E/S.

Revenue rose to $11.3 billion from $10.75 billion a year earlier, versus the average forecast of $11.03 billion.

"We weren't expecting fireworks for this quarter. I knew the company would control costs efficiently and there's a little bit of revenue upside," said BGC analyst Colin Gillis.

Networking vendor Juniper Networks yesterday admitted a Border Gateway Protocol (BGP) edge router issue hit a number of its customers, causing companies' websites to go down.

Customers across the world reported seeing core dumps by their Juniper routers on Monday, with many suspecting a BGP update was to blame. The vendor subsequently confirmed suspicions.

"This morning, Juniper learned of a Border Gateway Protocol edge router issue that affected a small percentage of customers," said Mark Bauhaus, executive vice president for services, support and operations at Juniper Networks.

A software fix is available, and we've been working with our customers to immediately deploy the fix.

"A software fix is available, and we've been working with our customers to immediately deploy the fix."

Juniper did not go into detail on what the "issue" was. It appeared a BGP update did not work perfectly with routers being managed by the company's Junos operating system, versions 10.2 and 10.3.

US-based ISP Level 3 Communications was hit hard, according to reports.

"Level 3's network experienced several outages across North America and Europe relating to some of the routers on our network," Level3 said yesterday.

"Our technicians worked quickly to bring systems back online. At this time, all connection issues have been resolved, and we are working hard with our equipment vendors to determine the exact cause of the outage and ensure all systems are stable."

Thousands of websites were reportedly affected. Link shortening site Bit.ly said it was experiencing issues thanks to Level 3's problems.

"bitly is operating normally, but there is a general connectivity issue at Level 3 affecting multiple ISPs and networks," the company said on its Twitter feed.

BlackBerry maker RIM was also hit by connectivity problems, alongside UK ISPs Easynet, Eclipse Internet and MerulaSupport, according to reports.

HP announced plans to stuff ARM-designed chips into its servers, as it looks to create super energy-efficient datacentres for cloud providers and large enterprises.

It marks a significant moment for ARM, which has done little to challenge the dominance of the x86 architecture used in the majority of servers today.

It also forms part of Project Moonshot a multi-year HP initiative designed to support those moving to "hyperscale environments."

HP Redstone is designed for testing and proof of concept. It incorporates more than 2,800 servers in a single rack.

HP is essentially looking to help companies truly consolidate their data centres, thereby making them more eco-friendly and efficient.

The company introduced the HP Redstone Server Development Platform, which will feature Calxeda's EnergyCore ARM Cortex processors. The system-on-a-chip processors should help companies basing deployments on the Redstone platform use 94 per cent less space in their data centres, HP said.

ARM launched its energy-efficient Cortex-A7 MPCore chip design last month with much fanfare around their power consumption capabilities. The British chip developer claimed its new processor design was five times more energy efficient than its predecessor, while also delivering significantly greater performance.

ARM won't be the only Redstone partner, however. HP confirmed future Redstone server deployments would come packed with chips from Intel and "others."

"HP Redstone is designed for testing and proof of concept. It incorporates more than 2,800 servers in a single rack, reducing cabling, switching and the need for peripheral devices, and delivering a 97 per cent reduction in complexity," HP said.

Select customers will be able to get their hands on Redstone offerings from the first half of next year.

To supplement the Redstone platform, the US tech giant launched the HP Discovery Lab. This services offering will let customers test out Redstone deployments before making any purchase. The first lab will open in Houston in January, with European facilities expected soon.

"The volume of data processed in financial markets has increased exponentially, and traditional scale-up or scale-out architectures are struggling to keep up with demand without vastly increasing cost and power usage," said Niall Dalton, director of high-frequency trading at Cantor Fitzgerald, a company looking into the Redstone technology.

"HP is taking a holistic approach to solving this problem and working to bring unprecedented energy and cost savings for tomorrow's large-scale, data-intensive applications."

Las Vegas may be best known for the glamour of its ostentatious casinos, but behind all the fun there is some serious IT hardware handling the torrent of gambling transactions that flow through all those casinos.

IT Pro managed to snag a fascinating behind-the-scenes tour of the Cisco-powered data centre behind it all.

Switch's intimidating private physical security force patrol the perimeter on foot and in Humvees, all armed with assault rifles.

Data centre company Switch operates seven data centres in the Las Vegas area, but SuperNAP is by far the largest with 400,000 square feet of servers using around 100 megawatts of power. Although purpose-built (retro fitting existing buildings is apparently very expensive) it resembles a non-descript series of warehouse-like structures. However, it's immediately obvious SuperNAP is no ordinary collection of buildings due to Switch's intimidating private physical security force. Comprised mostly ex-US Marines, they patrol the perimeter on foot and in Humvees, all armed with assault rifles.

SuperNAP's security office - the imposing rack of assault rifles is just out of frame.

To gain entry to SuperNAP we had to surrender our passports and the armed guards were almost never out of sight during our entire visit. We were also warned that any attempt at unauthorised photography would result in immediate expulsion from the premises.

The strict security is due to the organisations that use SuperNAP to store and process their data, from government agencies to financial services and technology companies including eBay, Mozy, Cisco itself and Global Cash Access (GCA). GCA, based in Las Vegas, handles almost all the financial transactions in the city's casinos, including on-site ATMs, and even provides those casinos with an Equifax-like service for judging the credit worthiness of a potential casino patron. According to the company's own figures it handled 90 million transactions worth a total of US$18 billion in 2010. GCA's clients include not just Vegas casinos, but gaming institutions in Atlantic City, Macau and the UK too.

All of this means that GCA not only has a major technological and contractual obligations to its casino clients, it is also subject to regulations and compliance rules from several jurisdictions. This is why GCA chose SuperNAP - according to Vice President of Information Technology Jim King only Switch could provide the necessary service guarantees his company required.

GCA currently uses 120 dedicated physical Cisco blade servers at SuperNAP all wired together using Cisco Nexus 7000-series switches and 7201 routers with external connections to GCA's offices using a bonded 2 Gigabit Ethernet backbone.

Just a small fraction of the servers available at SuperNAP.

All backups are handled using an EMC disk-to-disk system with no tape involved. Despite the presence of armed guards and mesh cages separating us from the servers, we weren't allowed to carry liquids of any quantity to prevent them from getting wet even the fire suppression system is waterless.

Although SuperNAP easily has the capacity to scale to even more servers to accommodate GCA's needs, including the provision of several cloud services or even migrate operations to another site through a dedicated loading bay, the transaction processing company has plans to reduce its server numbers by as much 60-80% using virtualisation.

That's a lot of cabling.

Even if that VMWare-enabled rationalisation happens, ensuring sufficient cooling and a consistent power supply for all that infrastructure will remain a primary concern for GCA and Switch. Instead of using an off-the-shelf cooling system Switch designed its own the Thermal Separate Compartment in Facility architecture or t-scif.

The externally visible parts of the cooling system at SuperNAP.

The external components resemble a series of large truck trailers on stilts and feed cold air into the server rooms and pump out all the hot air using large ducts protected by keypad-locked doors. T-scif uses one of four cooling methods - direct expansion, chilled water and direct or indirect evaporative.

Even though Las Vegas's desert location ensures predictable and consistent climatological and geological conditions, each truck trailer contains a self-contained weather station so that the system can choose the most appropriate cooling method depending on weather conditions. The concern over cooling even informed the colour of the LED lighting illuminating the walkways between the racks and racks of servers red and blue LEDs apparently emit the least heat (and match Switch's corporate colours incidentally). Switch prides itself on its claim that for every 1 kw of energy its systems use, it only needs a quarter of a kw of energy to cool them.

It's red and blue for a reason because it's cool.

Electricity supplied by the public grid is piped through what is apparently a specially designed, redundant distribution system. Despite the presence of the Hoover Dam hydroelectric facility in Nevada, almost all of that output goes to California, so the public grid in Las Vegas derives 60% of it supply from natural gas-fuelled power plants.

In the event of total power loss, Switch has fifty 2.8 MW diesel generators onsite. The company apparently has contracts with several fuel suppliers for diesel replenishment every eight hours and if that isn't enough, due to Switch's government clients, the security force is apparently empowered to requisition fuel wherever it finds it in the city.

Due to Switch's government clients, the security force is apparently empowered to requisition fuel wherever it finds it in the city.

These redundancies were of great importance for GCA in the event of a hypothetical 30-minute outage up to 12,000 casino transactions would otherwise fail. Switch has apparently investigated alternative power sources such as solar power and the possibility of recycling the heat generated by its equipment into energy, but neither methods are cost-effective or space-efficient enough at the moment.

The 'spine' of the redundant power distribution network at SuperNAP.

Intriguingly Switch claims that it can shave 25-40% off its customer's internet connection costs due to a project it inherited from Enron before that infamous company went under. Enron had planned to create a bandwidth marketplace so bandwidth could be traded like any other financial instrument or commodity. Although the marketplace never went live, the infrastructure powering it was built by Switch.

Switch's customers include business large and small, even customers who need just a single server cabinet, and services can be provided in as little as 25 days following contract signing. Pricing is greatly dependent on the customer's needs. Even if you think SuperNAP is a bit over the top, there's no denying the facility is impressive and an ever bigger data centre is on the way - SuperNAP II.

HP has announced a selection of technologies and programmes designed to help smaller businesses grow and compete more effectively with bigger players.

Part of HP's efforts include 40 new training centres to expand the HP Learning Initiative for Entrepreneurs (HP LIFE) programne. This programme has benefited from investments totaling $20 million thus far and has created 20,000 jobs in 6,500 new businesses since its inception in 2007, according to HP.

The company now plans to inject a further $4 million into the project.

"Organisations from 12 countries will each receive a grant of state-of-the art HP technology, financial contribution and professional support valued at approximately $60,000. The grant includes a professional train-the-trainer course and access to the HP LIFE network as well as innovative online and offline training tools," said HP spokeswoman Elisa Greene in a blog post.

"HP Learning Initiative for Entrepreneurs (HP LIFE) is a global training program that trains entrepreneurs and small businesses owners to develop essential IT and business skills to grow their businesses... The programme is committed to enabling microenterprises, small and mid-size businesses, (SMBs) to create new revenue streams, promoting growth and opportunities in their local communities."

A range of other solutions focused on data storage, networking and collaboration were also announced by the company, with the aim of giving smaller businesses "optimised performance at a lower cost, simplified deployment of services and reduced management costs."

This new tech comprises a number of entry-level servers and switches optimised for rack deployment, including the ProLiant ML110 G7 and the ProLiant DL120 G7 servers, as well as the V1810-48G and V1410 switches.

One other area that will no doubt prove attractive to small businesses is the HP Business Risk Mitigation (BRM) service. This has the potential to reduce backup and recovery times by 90 and 87 per cent respectively, according to the tech giant.

Today saw the merger of Orange and T-Mobile take the next step as its customers can now access both networks.

The newly formed company, called Everything Everywhere, now provides mobile network access to 27 million people in the UK and dominates with market share.

Whilst customers will need to register online to access the previous rival network, the extra coverage will be available for free and no changes to user's tariffs will be needed.

Tom Alexander, chief executive (CEO) of Everything Everywhere, said: "Today's switch-on is the culmination of a unique and hugely complex technical project. But the result is simple our customers now get two networks for the price of one."

"That means 27 million consumers can now keep close to the people, places and things that matter to them in more places than ever before. This is the first step in our vision of giving our customers instant access to whatever they want, wherever they are instant access to everything, everywhere."

At the moment, it is only the network for voice and text being shared, but the plan is to extend this to 2G, 3G and Wi-Fi connections in the future. The company has also claimed next year will see customers able to choose between T-Mobile and Orange's coverage for whichever offers them the strongest signal.

If you are a T-Mobile customer and want to register, click here, for Orange customers, click here.

HP introduced on Wednesday a host of networking and storage products aimed at SMEs and based on the integration of voice, messaging and collaboration offerings.

The first products, the V1410 layer 2 gigabit switch family forms part of the V-series networking portfolio, and comes in eight, 16 and 24-port versions. The V1410 products are entry-level products each with two mini-GBIC slots. One has the option of fibre connectivity and all are capable of working without a fan. They can be wall-mounted or sit on a table-top.

HP's StorageWorks X310 Data Vault is a storage offering that offers deduplication and back-up features, providing local and remote file access for up to 10 PCs or Macs. The StorageWorks X1000 Network Storage System, meanwhile, offers up to twice the number of drives of prior generations in half the space.

The HP StorageWorks P2000 G3 SAS (serial-attached SCSI) Modular Array provides twice the capacity of the previous generation. It features 6Gb SAS technology, providing shared storage consolidation without the extra networking cost. It does not require the skills traditionally needed to run a storage area network, HP said.

The Unified Communication & Collaboration offering - which comes from a partnership between HP and Microsoft - combining servers, storage, networking and software which the companies intend will help to streamline firms' messaging, video and voice applications.

HP Virtual Rooms offers an interactive workplace, with a range of monthly billing options, while HP 48Upper is an online service that is designed to make the management of IT processes easier. This could include project planning, operations and service management, HP said.

HP also introduced a Download Store, featuring six pay-as-you-go support and collaboration services. One is named HP SmartFriend Assistance, where HP provides direct professional help in areas like software installation and printer issues. Another is the HP Performance Tune-Up Service, which offers support for users with poorly-performing machines.

CASE STUDY

It has been a great season for Fulham FC, with a decent finish in the Premier League, manager Roy Hodgson receiving an abundance of praise and now a Europa League final.

As the team gets ready in Hamburg for the biggest game in the club's history, one man going over to cheer them on will be Fulham's IT manager Nicholas Pendlebury.

IT PRO caught up with the man himself to talk about why it isn't just on the pitch where the club has been enjoying success in recent times.

Out with the old

In both football and IT, when things are not working, they need changing. For the latter, the cost of not making alterations to the infrastructure can prove disastrous. Fortunately for Pendlebury and his small team, the problems were noted and suitably addressed in time.

The area that the IT manager was experiencing problems with was the local area network (LAN) and Ethernet switches in particular. Around a year and a half ago, the club had switches from a variety of vendors, meaning there was no standard across the board and the IT team was "very limited" with what they could do, Pendlebury explained.

The switches in place were not capable of delivering the technologies that the organisation wanted to help improve performance, benefit customers and reduce costs.

Pendlebury was also seeing increasing demand for LAN bandwidth from the ticket turnstiles, television networks as well as media commentators.

Both the capacity and resiliency of the network had to be enhanced so that match days, when traffic is significantly higher, went smoothly. Video and data traffic crossing the LAN needed to be prioritised as well so that visiting media companies were given a top service.

And it was not just Craven Cottage that needed attention, as the Cottagers' training facilities required an improved network infrastructure as well.

In with the new

It was for these reasons that Pendlebury and his team decided to bring in a new LAN infrastructure deployment using Juniper Networks EX3200 and EX4200 Series Ethernet Switches.

"We went to tender, we ended up with two vendors, with Juniper being one of them. We chose Juniper due to the switch specifications, [which offered] more than anyone else," Pendlebury explained. He also pointed to the power consumption of the switches, which the IT manager said were lower than anything else on the market at the time.

"A key driver for me was the single Junos operating system across all the different switches we have, whereas the other vendor had three different operating systems for their solution. So, it is a lot easier for myself to manage the things that we do manage."

Pendlebury also called on Scalable Communications, Juniper's preferred partner and a convergence solutions provider, to help implement the new infrastructure. The tech was rolled out at both the training facilities and the ground in just two days.

Results on and off the pitch

Following the introduction of these services, the IT operations at Fulham have been especially fluid, much like the team's smooth passing game on the pitch.

While it used to take more than two hours to produce a video replay of the match for player analysis, it now takes only 15 minutes. Match days, meanwhile, are considerably less stressful for Pendlebury and he claimed there has not been a single failure since the Juniper equipment was installed.

Since the implementation of the new infrastructure, the team has also achieved an average cost reduction of 30,000 a year, which had gone on operating and maintenance costs for the old system, according to Pendlebury. "For us, this was a massive saving," he explained.

"The new system has given us a lot more stability, which we didn't have. We were having a lot of network problems and downtime. It's made a huge difference."

Fulham's IT future

The change in switches has also enabled Pendlebury to look into other areas for future development. "We are now looking to implement IPCC TV, which will give us 160,000 saving over three years from our current system," he said.

He is also looking to bring in a new Wi-Fi network with power over Ethernet (POE). The current Wi-Fi is somewhat dated and other POE technology is needed to help power it, Pendlebury explained.

Fulham FC is close to achieving PCI Compliance in the near future as well, Pendlebury said, and the new Ethernet switches have been essential in this. "There are a lot of changes you have to make securing data, monitoring and Juniper has helped us with the tools it offers in its suite to achieve compliance."

IT the great enabler

IT is a key enabler for business success in numerous ways and football clubs are no exception, with Fulham being a case in point. Leaving systems alone can result in a plethora of problems and the alterations at the club have shown how being brave and making changes can provide two of the most valuable business commodities around: time and money.

Whether the team can defeat Athletico Madrid tonight or not, it can rest assured that the IT set-up at the club will have its back for the foreseeable future.

Nvidia has unveiled a new technology for laptops which will automatically switch from integrated to discrete graphics depending on the task.

While integrated graphics chips tend to be quite gentle on battery life, they don't have the muscle to tackle more intensive tasks such as gaming. In contrast, discrete graphics cards tend to offer more grunt at the cost of a higher power draw - putting a significant strain on battery life.

In an effort to offer customers the best of both worlds, laptop manufacturers have begun selling laptops offering both an integrated chip and a discrete graphics card. However, they've been required to manually switch between the two, and according to figures presented by Nvidia, less than one per cent of people with the feature take advantage of it.

In an effort to address this Nvidia's Optimus promises to automatically flip on the discrete graphics card when the user requires the extra power - running a game, for example. However, the moment they go back to writing an email, or browsing the web, Optimus will turn the card off and hand control back to the integrated chip.

One potential pitfall could be that Optimus depends on software profiles to tell it whether a task requires discrete graphics or the integrated chip. For applications that aren't recognised by Optimus, users will be able to activate the correct option with a simple right click, or by creating their own profile through the Nvidia Control Panel.

Nvidia is promising to constantly update the Optimus software with new profiles, meaning users shouldn't be tripped up too often.

The graphics card company said the technology will work with a range of Intel processors, including the Pine Trail platform, and will be integrated with its own GeForce 200M and 300M series, and next-generation Ion platform.

The first Optimus-enabled laptops will arrives courtesy of Asus at the end of February, with models from other manufacturers debuting in the next few months.

Smartphones and other devices could soon become pressure sensitive, as a Samsung components arm has licensed quantum-based technology from a Yorkshire firm.

Samsung Electro-mechanics has licensed the Quantum Tunnelling Composite (QTC) technology from Yorkshire-based Peratech, in order to create more sensitive devices. QTC uses spiked bits of nanoparticles in polymer to create the effect.

Peratech's switches will go into navigation keys - called Navikeys - on mobile phones, offering continuous instead of the standard on/off input in standard switches. That means the switch will know how much pressure is being used at any given time.

"This adds a whole new dimension to the input device and this proportional input opens up a raft of new ways to interact with the phone," Peratech explained. "For example, the speed of scrolling through a list or the speed of movement in a game can vary depending on how hard the switch is pressed, making interaction faster and more intuitive."

Peratech joint chief executive Philip Taysom said we can expect to see phones with such tech soon. "Samsung EM supplies components to most of the leading phone manufacturers so our technology will soon be used across a wide range of next generation phone models," he said in a statement. "In fact, a Navikey using QTC from Samsung EM is already being used into a Tier 1 mobile phone."

However, a spokesman for Peratech told IT PRO that the firm wasn't allowed to disclose which phone that was. "It's on the streets but we're not allowed to tell you," he said.

The Peratech site does have an "anonymised" image of the phone in question.

With ever-growing numbers of businesses and users, quite simply the internet is running out of Internet Protocol addresses with which to create new website domains.

With the problem compounded by the need for IP addresses for mobile phone users, the twenty year-old IPv4 system allocating IP addresses is showing the strain. IPv6 is the newest version of the Internet Protocol offering space for longer addresses and increasing the number of available IP addresses from 4 billion to a staggering trillion trillion.

According to the latest IPv6 Deployment Monitoring Survey sponsored by the European Commission, not enough businesses are switching over to IPv6. The survey, which included 610 respondents from 54 countries, revealed that only 17 per cent of companies are making use of IPv6.

The position has only marginally improved since ARIN's survey in March 2008 with businesses remaining uncertain about the benefits of switching over. Of those businesses that weren't considering switching to IPv6, almost 70 per cent couldn't understand the need for change.

The survey also revealed that 92 per cent of ISPs are not using IPv6 and are themselves unsure about whether to promote IPv6 uptake to their customers.

Why switch to IPv6?

So why should businesses change over to IPv6? Even RIPE's "IPv6 Act Now" site only warns us that "failure to adopt these new resources could mean a slowing in the pace of internet innovation" which isn't something most businesses are likely to worry about.

IPv4 addresses are likely to continue working long after all the available addresses run out, so the problem itself is only likely to affect new businesses or current businesses looking to expand their web presence.

That said, a change to IPv6 seems inevitable. Cisco's view is that "every customer in the market will face the inevitable transition from IPv4 to IPv6."

As such, the change to IPv6 will eventually be forced upon businesses as demand for IP addresses exceeds supply along with the associated costs of retraining staff, enhancing management tools, routers and operating or deploying IPv6-enabled versions of applications.

Microsoft have already built support for IPv6 into Windows from XP to 7, so it really boils down to a question of whether to change over to IPv6 sooner or later?

What are the advantages of switching?

Besides the capacity to have longer IP addresses - 128 bits instead of the current 32 - which might prove beneficial for various web apps, IPv6 offers tougher security thanks to the core feature of IP(Sec) which secures B2B data flows using AES and Triple-DES encryption.

IP(Sec) also offers strong authentication with a header extension that checks each data packet comes from the source address it claims and makes use of cryptographic keys to establish a session between computers or servers.

Besides better security, IPv6 also comes with auto-configuration, with IPv6 hosts detecting the presence of local networks and routers automatically potentially saving businesses money in terms of IP and network management. Auto-configuration will also aid wireless mobile devices in connecting when travelling and discovering foreign networks.

Multicasting, or sending data packets to multiple recipients, is also supported by IPv6, which is handy for videoconferencing and streaming video to other business users. While IPv4 also supports multicast, it's optional and not every router or host currently supports it. IPv6 also has an anycast mode for sending information to the nearest member of a group of recipients.

There are other differences too. IPv4 treats all packets of information the same whereas IPv6 implements quality-of-service (QOS) by classifying packets using traffic class and flow label fields essentially this means data flows can be prioritised and requests for special handling can be made. Finally, IPv6 is more customisable than IPv4.

It's possible to create extension headers that add more functionality so with any luck we won't be facing the same need for migration in the future.

Timescale for action

When will the supply of IP addresses actually run out? Estimates vary, but most suggest that the Internet Assisted Numbers Authority's (IANA) unallocated addresses will be exhausted sometime between 2011 and mid-2012.

The European Commission feels a "significant step" should be taken as early as 2010. RIPE's site estimates that 88 per cent of IP addresses are already used up.

Certainly the business world doesn't seem on course currently to meet the European Commission's IPv6 target adoption rate of 25 per cent by next year.

Despite the EC's calls for Governments to lead the way in switching over to IPv6, to which the Internet Society has added its voice, the UK government has yet to intervene, although it has recently released specs for IPv6 compliance required for government network procurements.

Advice on switching over

Of course, despite the obvious benefits of IPv6, switching over is likely to mean costs for business in retraining staff, enhancing routers and deploying IPv6-enabled versions of applications.

Probably the biggest issue for businesses is integration. IPv4 addresses and IPv6 addresses can't communicate directly with each other but they are likely to co-exist together for a long transitional period.

The most common technique for integration is likely to be using a tunnel broker tool capable of providing IPv6 functionality via IPv4 infrastructure on the web.

The first step for businesses thinking of taking the plunge to IPv6 is to find out whether their computers and devices have IPv6 support - a full list is here.

If you are thinking of migrating over to IPv6, consider drawing up a migration plan. There are plenty of sites on the web to help you including IPv6.com, as well as ones from Cisco and Lumeta.

Cisco has launched a new switch product for its Nexus line designed to integrate with blade servers.

Details are still sparse as Cisco has yet to confirm release dates and who it will be partnering with.

The Nexus 4000 is built to fit in a blade server system and enables 10 gigabit ethernet and fibre connectivity through one switch, negating the need for separate products and extra wiring.

The switch will support the NX-OS operating system but the key is that it will fit into non-Cisco chassis' from the likes of HP, IBM and Dell.

The company claims the switch will be high performance with a 1.2 microsecond delay across the chip and 1.3 to 1.5 microseconds from port to port.

HP is offering its customers SSDs to put into its ProLiant G6 server range.

The company is teaming up with Samsung to offer the upgrades that will significantly increase power and cut down on power consumption.

Jim Ganthier, vice president of marketing for Industry Standard Servers at HP, said in a statement: "Customers demand fast, energy efficient servers that enable them to do more with less, cutting the costs associated with running their data centres."

He added: "Samsung's latest SSD technology coupled with HP ProLiant servers delivers energy efficient server platforms to enable customers to slash their power usage and reduce costs."

Samsung claimed that by replacing a 10,000 or 15,000 rpm HDD with single-level-cell (SLC) 3Gbps SATA SSDs a server can run 40 to 50 times faster, depending on how they are implemented.

"Using SSDs as the primary storage medium in enterprise servers will provide optimal value for data centres, with their exceptional low-power attributes, long-term reliability and outstanding performance," added Jim Elliott, vice president of memory marketing at Samsung, in a statement.

The SSDs are also a lot smaller so could free up physical space within the server, which could be used to put in more DRAM. They can replace between four and 40 15,000 HDDs in a server array.

HP has also announced the upcoming launch of two new networking switches, the 1Gbps HP ProCurve 6120G/XG Blade Switch and the 10Gbps 6120XG Blade Switch.

The switches can be incorporated into existing blade server infrastructures and the company claimed they will help with overall data centre management.

The two switches will be available from 1 October.

Visit IT PRO's sister title Channel Pro for news, in-depth analysis and industry comment aimed at distributors, resellers and anyone else working in the IT channel.