The EdgeLine EL8000 ups HPE’s game, delivering a high performance edge compute solution in a compact and rugged 5U half-width chassis. Naturally, 5G adoption is a key focus for HPE but it has many other target markets for the EL8000 including retail, manufacturing, transportation, defence and energy production.

Whereas the EL1000 and EL4000 systems employ HPE’s old Moonshot Xeon D and Xeon E3 server cartridges, the EL8000 gets a brand new purpose-built server blade. It supports up to four ProLiant e910 blades, making the EL8000 one of the first solutions to deliver Xeon Scalable computing to the network edge.

HPE EdgeLine EL8000 review: Chassis design

The EL8000 is designed for harsh environments – it’s shock and vibration resistant and rated for continuous operations in temperatures between 0 and 55 degrees C. The 5U chassis can be deployed as a standalone unit or rack-mounted, where you can place two alongside each other and even a third sideways on behind them.

The four upper 1U slots each accept an e910 blade and if you want more options, you can mix them with 2U versions which double the number of PCI-E slots. All system cooling is handled efficiently by an array of nine hot-plug cooling fan modules at the rear.

Two 1,500W hot-plug PSUs slot into the lower 1U space and alongside there’s room for up to two unmanaged 10GbE switch blades. Tucked up in the lower right-hand slot is HPE’s ECM (EdgeLine chassis manager) module which provides web browser access for managing the chassis, monitoring the blades and controlling power.

HPE offers a range of casing options for the EL8000; a 19in wall-mount enclosure allows two to be placed side by side, or you can go for a shipping container so the system can be delivered and operated without unpacking it. Our review unit was supplied in a battle-hardened aluminium wheeled case and it can be protected out in the field with a rugged, water-resistant version.

HPE EdgeLine EL8000 review: The e910 blade

The ProLiant e910 server blades pack a lot into their compact dimensions, with the single Xeon Scalable CPU socket located at the rear and flanked on each side by six DIMM slots. Further forward are two PCI-E expansion slots plus a dedicated slot in the middle for an optional mezzanine card.

The blades have an embedded Gigabit port which can be accessed from the ECM, as it uses its second Gigabit maintenance network port to aggregate all four connections. More importantly, they each have an Intel X722 quad 10GbE backplane connection where two can be routed through to the unmanaged network switch blades which terminate them on a 10GbE SFP+ port.

If you install the quad 10GbE mezzanine card, its QSFP+ port is presented at the front of the blade. This doesn’t go through the switch module though, so you’ll need to use HPE’s direct-attach breakout cables to split it into four external 10GbE ports.

The blades only support M.2 SATA or NVMe SSD storage, and have two on-board 2280 slots. This can be easily trebled with a riser adapter that fits over the top of the mezzanine card and supports longer 22110 M.2 cards.

The 2U server blades are essentially the same, with the main difference being that they have an extra 1U upper unit which provides four PCI-E slots. The chassis also supports a 2U storage bay providing 8 SFF hot-swap bays plus Smart Array controller. This is available as a factory order and can only be used with one 2U e910 server blade.

HPE EdgeLine EL8000 review: System management

The blades each have an embedded iLO5 management controller with all of them piped through the ECM’s network port. It presents a smart web console packed with information about server blade status and essential details on critical components.

We did come across one glitch though; we were unable to use HPE’s Intelligent Provisioning tool to deploy Windows Server 2019. This tool expects to see a Smart Array controller and with NVMe-only storage available on our blades, it refused to proceed, requiring us to manually deploy the OS from a mapped ISO file and apply all necessary drivers.

The chassis management web interface is very informative and opens with a dashboard view of blade health status and an active graphic of the chassis with all installed components. More information is available on the blade summary page and each entry provides a direct link to their iLO5 interface.

Clicking on a blade entry takes you to its personal power controls so you can gracefully bring the blade down, reset it or recycle power. Usefully, you can select a one-time boot option without visiting the iLO5 interface.

Strong platform security extends to silicon root of trust fingerprinting for the chassis and blades so you know their firmware hasn’t been compromised. The ECM supports SSH access and maintains audit logs, while the iLO5 chip allows administrators to use SSO to manage user access to all blade features and offers four encryption levels - including the highly secure FIPS and CNSA modes.

HPE EdgeLine EL8000 review: Verdict

The EdgeLine EL8000 shows HPE is getting very serious about edge computing. This 5U system delivers a smart modular design, allowing it to be customised to suit specific scenarios and the e910 blades are amongst the first to put Xeon Scalable CPUs close to the edge.

Network options are extensive and the e910 blade design and features allow you to create a high performance solution capable of running many enterprise apps. Its rugged design allows it to go places other servers daren’t venture, and HPE offers a wide range of portable, battle-ready transport cases for ‘deliver and deploy’ services.

HPE EdgeLine EL8000 specifications

The launch of these new servers follows last year’s release of the ThinkSystem SR635 and SR655 single-socket servers.

The ThinkSystem SR645 and SR665 two-socket server platforms come equipped with 128 central processing unit (CPU) cores per system and up to two 64-core AMD EPYC 7002 processors.

“Our new Lenovo ThinkSystem servers are designed for workloads such as in-memory databases, advanced analytics, virtualization, and AI,” noted Kamran Amini, VP and general manager of server, storage and software defined infrastructure at Lenovo Data Center Group.

“With the exceptional power, speed, and onboard storage of these new servers, our customers have the ability to handle the increasing data requirements of today’s workloads with the scalability to grow with their business.”

Lenovo created these new servers with an eye on the financial, retail and manufacturing industries, which require faster transaction processing and improved data analytics. These industries also require optimal grid-computing capacity without compromising the efficiency or the total cost of ownership, these new servers aim to deliver in these areas too.

With the addition of the new ThinkSystem SR645 and SR665 servers, Lenovo can help its customers in these industries ramp up higher-performance workloads to improve efficiency.

With their higher core counts, customers can scale down the number of servers they need, making them ideal for areas where rack space is limited. The financial industry, where servers are typically housed in tight quarters near financial exchanges, is a great example where these high-performing servers will work best.

“Lenovo’s endless pursuit of innovations to accelerate our customers’ intelligent transformation has been paramount in our rise to become one of the fastest-growing data center OEMs in the world,” shared Doug Fisher, Lenovo Data Center Group COO, and SVP of business units.

Fisher continued, “Today Lenovo is expanding our AMD relationship with new, fully optimized solutions to help our joint customers address complex and data-intensive workloads, enabling them to do more with less while still providing uncompromised end-to-end security.”

The new Lenovo ThinkSystem SR665 and SR645 solutions are now available through Lenovo sales reps and partners worldwide.

Its two-socket blade server FusionServer E9000 CH121 V3 underwent three SPEC CPU2006 tests, SPECint_base2006, SPECint_rate_base2006, and SPECfp_rate_base2006, scoring 74, 1790 and 1120 respectively.

The FusionServer E9000 CH121 V3 uses converged architecture that integrates compute, storage and networking and is based on Intel Xeon E5-2600 v4 CPU, which can provide up to 22 cores and deliver 20 per cent higher performance than the previous generation E5-2600 v3.

According to Huawei, it is this new technology that has made its latest range of FusionServer 20 per cent more powerful than its predecessor.

The server also has all-flash blades that support up to 96 NVMe SSDs per chassis, hard disk storage blades that support up to 48 3.5in hard disks per chassis and board-level liquid cooling blades, which the company claims consume 40 per cent less power than the previous E9000 FusionServer range.

When fully configured with all-flash blades, it delivers up to 307TB storage per blade and 10 million IOPS per blade.

Zhang Xiaohua, general manager of Huawei's blade servers, said: "Huawei's blade servers have repeatedly set new records in the industry's most authoritative SPEC tests, proving Huawei's powerful server development capabilities."

"With continuous innovation, Huawei FusionServer E9000 blade server maintains its excellent performance and reliability, while always adhering to its openness philosophy," Zhang concluded.

While Huawei FusionServer E9000 now holds the two-socket blade server record, the current one-socket blade server record is still held by Oracle's SPARC T7-1 servers, which scored 1120 on SPECint_rate_base2006 and 801 on SPECfp_rate_base2006.

The HP Proliant m400, aimed at web server applications, is based on the X-Gene Server on a Chip (SoC) from Applied Micro Circuits.

The server sports a 64-bit ARM chip with eight cores running at up to 2.4GHz. The server comes with Canonical's Ubuntu Linux as well as cloud orchestration tools Juju and Metal-as-a-Service (MAAS) software for automatically provisioning bare metal servers.

The server is designed to save power, cooling and space with HP claiming a 35 per cent reduction in total cost of ownership in comparison to normal rack servers.

HP touted Sandia National Laboratories and the University of Utah as early customers for the m400.

"By giving researchers access to the HP Moonshot platform in our CloudLab facility, we empower them to develop fundamental breakthroughs that have the potential to change the capabilities, performance, reliability, and security of future clouds," said Robert Ricci, research assistant professor of Computer Science, University of Utah.

The second new server is the m800, which sports a 32-bit ARM processor from Texas Instruments which comes with four ARM Cortex A-15 cores, and integrated digital signal processors (DSPs). These servers can be linked together using HP's 2D Torus Mesh Fabric in combination with Serial Rapid I/O.

One of HP's customers already using the m800 is PayPal in its Systems Intelligence project.

"The HP ProLiant m800's combination of ARM and Multicore Digital Signal Processors with high-speed, low-latency networking and tiered memory management creates a very energy efficient, extremely capable parallel processing platform with a familiar Linux interface," said Ryan Quick, principal architect of the Advanced Technology Group at PayPal

Analysts said ubiquitous cloud-enabled smart devices are a driving force behind this shift in IT infrastructure.

"Service providers deploying context-rich services to these devices are building massive new datacenter capacity and looking to their vendors to optimize infrastructure for their specific workloads," said Patrick Moorhead, of Moor Insights and Strategy in a whitepaper published by HP.

Pricing for the 64-bit HP ProLiant m800 starts at $81,651, while pricing for the 32-bit HP ProLiant m400 starts at $58,477.

The launch happened at this year's CeBIT conference in Hanover, Germany, with the company also saying the solution cuts 75 per cent of the mechanical costs associated with cooling servers.

Cost savings on the maintaining side might be plentiful, but the set-up price is no drop in the ocean

The product uses 3M's Novec liquid as a coolant, which Peter Hopton, founder and chief technology officer (CTO) of Iceotope, called the "environmentally friendly little brother" of Fluoroinert 3M's previous coolant incarnation.

A water jacket containing Novec wraps around components and travels through the servers in fast currents at a rate of centimetres per second. With its low surface tension, the coolant "gets into all the cracks" to absorb the heat from servers and take it away to heat exchanges, where the liquid can either be cooled or used to heat other buildings and water within the complex.

The technology can not be incorporated into existing servers, however, and businesses would have to buy new servers or modules as Iceotope refers to them to install the technology.

Cost savings on the maintaining side might be plentiful, but the set-up price is no drop in the ocean.

Customers can buy cabinets, featuring six server modules which look like blades as well as the pump and heat exchanges, for 19,995. However, the cabinets can house up to 48 modules, which start at 3,995 for fully configured servers containing two six core Xeon E5 processors, 64GB RAM, 40Gb Infiniband and SSD storage. Users are able to choose between AMD and Intel processors.

Hopton told IT Pro the technology could soon roll-out across the data centre, with products for switches, routers and GPUs in the pipeline. He defended the upfront prices though, saying the overall benefits would save a lot more money in the long term.

"No chiller equipment is needed so when you are kitting out a new data centre, you don't need to buy it," he said.

"It is not massively expensive and if you need new servers it is cost effective when you start to look at the money saved on energy consumption."

Although he couldn't confirm any UK customers yet, Hopton did say it was being implemented within UK data centres and the test company would be revealed soon.

The chip manufacturer has always come second in the server space against major rival Intel, but hopes the new addition to its portfolio will bolster its position and get more AMD chips into cloud environments.

AMD specifically wants to help companies with web content, social networking, search and video workloads and aims to reduce complexity within the data centre, lower cost and energy consumption and increase performance.

By acquiring SeaMicro, we are accelerating AMD's transformation into an agile, disruptive innovator.

"By acquiring SeaMicro, we are accelerating AMD's transformation into an agile, disruptive innovator capable of staking a data centre leadership position," said Rory Read, president and chief executive (CEO) of AMD.

"The unmatched combination of AMD's processing capabilities, SeaMicro's system and fabric technology, and our ambidextrous technology approach uniquely positions AMD with a compelling, differentiated position to attack the fastest growing section of the server market."

SeaMicro claims its servers use only 25 percent of the power of traditional servers, whilst only taking up one sixth of the room in a data centre. It also touts 12 times the bandwidth per core.

Andrew Feldman, CEO of SeaMicro CEO, added: "SeaMicro was founded to dramatically reduce the power consumed by servers, while increasing compute density and bandwidth."

"By becoming a part of AMD, we will have access to new markets, resources, technology, and scale that will provide us with the opportunity to work tightly with our OEM partners as we fundamentally change the server market."

Feldman will become general manager a new Data Centre Server Solutions company at AMD to incorporate his firm's technology.

The EC said there had been a "relative decline in HPC use and capabilities," but it hopes that will be reversed with a doubling of investment in supercomputing.

High Performance Computing is a crucial enabler for European industry and for more jobs in Europe.

The EU plans to increase investment in the industry from 630 million to 1.2 billion and run machines that can perform 1,000,000,000,000,000,000 tasks per second, otherwise known as exascale computing.

"High Performance Computing is a crucial enabler for European industry and for more jobs in Europe," said Neelie Kroes, the EC's vice president responsible for the Digital Agenda.

"It's investments like HPC that deliver innovations improving daily life. We've got to invest smartly in this field because we cannot afford to leave it to our competitors."

The EC wants to see greater use of HPC systems and services amongst industry players and SMEs.

HPC has traditionally been reserved for larger enterprises, due to the significant cost. Supercomputers can cost as much as 100 million or more just to build.

Major manufacturers do lease out usage of their HPC machines, however. Indeed, Intel told IT Pro last year that supercomputing was going to become much more mainstream in the coming years.

The EC said it would like to see "centres of excellence" established for software in scientific fields like energy, life-sciences and climate.

Andy Grant, an HPC expert at IBM, welcomed the announcement.

"It's not just good for us but good for the European economy," he told IT Pro.

"It's good that the EU is recognising importance in HPC. It's something the UK Government has been doing as well."

As for whether HPC capabilities could reach SMEs, Grant claimed there has been a lot of effort to make HPC more SME friendly.

"We can't say for certain the barrier for entry will come down but I would advise SMEs go to cloud providers and see if they can dip their feet in HPC," he added.

Supplied by Broadberry Data Systems, the Intel Modular Server is designed for SMBs looking to consolidate their IT services onto a single platform. It's designed to beat the blade competition on value and this latest version supports Intel's integrated virtualisation features.

Intel's Modular Server is designed for SMBs looking to consolidate their IT services onto a single platform.Size-wise, the Modular Server matches the HP and Fujitsu models with its 6U chassis. Server blade support is more limited though, since while it has room for up to six compute blade modules, Intel only offers one model.

The modules use Intel's custom MFS5520VIR motherboard, which supports two Xeon 5500 or 5600 processors and up to 192GB of memory. Both the Fujitsu BX400 and HP c3000 have eight half-width slots and provide a wider range of server blade options.

When it comes to storage, the two fixed bays at the front of the Modular Server each support three LFF drives or seven SFF drives. This is where Intel differs radically from the competition, as neither the BX400 nor c3000 have any integral storage support, and so the server blades must carry their own or slots must be dedicated to storage blades.

All drive bays in the Modular Server are linked to the chassis midplane and routed to a storage controller blade at the rear. This allows an internal SAN to be set up where selected disks are placed in storage pools that can then be used create virtual drives, each with their own RAID array.

The storage controller supports all key RAID arrays, including RAID-6, and Broadberry offers a choice of SATA, SAS, near-line SAS and SSD drives. Storage pools can contain multiple virtual drives and each is assigned to a module that sees it as local storage, so the modules themselves are completely diskless.

The storage arrangement provides good redundancy if a module fails, its virtual drive can be reassigned to a spare module and booted as usual.This arrangement provides good redundancy if a module fails, since its virtual drive can be reassigned to a spare module and booted as usual.

The storage management blade also has a SAS expansion port for adding external storage, but note that only 3Gbit/s speeds are currently supported. Assigning storage is straightforward, too a flowchart graphic shows the content of each storage pool and which compute modules the virtual drives are assigned to.

A graphic shows clearly how the internal SAN is arranged and which virtual drives are assigned to each compute module.

A management I/O module provides full remote access to the server chassis and its components. Switch options are basic, as the chassis only supports a maximum of two 10-port L2/L3 Ethernet switch blades. The compute modules do have a 10Gbit Ethernet internal connection to the network midplane, but a 10Gbit Ethernet external switch module isn't currently available.

Intel's Modular Server Control (MSC) browser interface offers good levels of access to all components. A dashboard shows the status of the chassis, compute modules, power supplies, drives and system health, while a row of tabs above provides views of the front and rear panels, along with installed components.

Intel's MSC provides good remote access to the server and plenty of operational information.

The BIOS and operating system of each computer module can also be viewed and controlled remotely via KVM over IP, and this allows modules to be switched on and off individually, too.

Virtualisation management is enabled with an activation key and Intel has plumped for the open-source KVM (Kernel-based Virtual Machine) software. This is classed as a hosted hypervisor', since it runs in Linux.

Virtualisation setup is nicely automated simply select the Virtualize option next to a compute module and wait it to be preparedConsidering it only costs 495 per chassis to activate, KVM provides a goodly range of features. Up to 256 virtual CPUs per compute module are supported (up to eight per VM) with up to four virtual NICs.

Virtualisation setup is nicely automated simply select a compute module, select the "Virtualize" option in the menu alongside it and wait for the module to be prepared. A VM storage pool can be created and assigned while this is in progress, but note that a minimum of four SAS drives or SSDs are required and dual active/active storage controller modules are not yet supported.

This set-up process also creates a Cluster Share storage area for copying ISO images to and, when defining VMs, these can then be used for boot purposes.

Once the modules have been virtualised, a wizard helps create your VMs.

KVM's live migration feature also allows a VM to be moved from one compute module to another, without any downtime. The migration process is simple just drag and drop the VM icon from one module to the other in the Modular Server Control interface and there are two options.

VM Live Migration is a simple drag and drop manoeuvre which worked perfectly during testing.

Temporary migration is particularly useful for maintenance tasks, since it only runs an image of the VM in the destination module and reverts back to the original module when the duplicate is closed down. Permanent migration, on the other hand, moves everything over to the destination module.

We tested KVM's migration feature by creating a VM hosting Windows Server 2008 R2 on one module and left a second virtualised module running with no VMs set up. Prior to running a temporary migration, we opened a remote desktop (RDP) session and pinged the VM from another system.

The temporary migration took less than 30 seconds and at no time did we lose contact with the hosted OS. We then ran a permanent migration that took 50 seconds, also with no interruption to services.

The main competitor for the Modular Server is the HP BladeSystem c3000 and Intel is a clear winner on price. The lack of support for 6Gbit/s SAS speeds and 10Gbit Ethernet switch blades are drawbacks, but the new virtualisation features are good value, easy to use and make the server even more flexible.

Verdict

Broadberry’s Intel Modular Server is more cost-effective than HP’s C3000 and just as easy to deploy. Its internal SAN allows all server blades to be diskless, although the controller still only supports 3Gbit/s speeds. The virtualisation option is excellent value and its live migration worked seamlessly during testing, but bear in mind that support for dual storage controllers has yet to be implemented.

Chassis: 6U rack enclosure with six compute module slots Power: 2x 1000W hot-swap supplies (max 4) Fans: 2x hot-swap Storage 1x Controller Module max 2), 6x internal 3Gbit/s SAS ports, 1x external SAS port RAID support: RAID 0, 1, 1E, 10, 5, 6 Storage: 4x 900GB Seagate Savvio (10K rpm). 5 SFF 6Gbit/s SAS hot-swap hard disks (max 14) Switch module: 1x 10-port L2/L3 Gigabit Ethernet (max 2) Management: Management blade with 10/100 Ethernet, serial port Extra Features: Virtualisation key with one year support, Shared LUN key Two Compute Modules each with the following: CPU: 1x 2.4GHz Xeon E5620 Memory: 8GB 1066MHz DDR3 (max 192GB) Storage: Embedded LSI SAS1064e SAS Network: 2x Gigabit Ethernet (max 4) Warranty: 3-year on-site (next business day)

At the company's launch of its new ProLiant Gen8 server range, the focus was on automation - removing manual tasks from the datacentre and taking away complex systems sucking up most of IT administrator's working hours.

However, with less need for administration in the datacentre, some believe there could be less need for IT admins themselves.

IT admins are celebrating. They are celebrating that they don't have to spend all their time keeping their data centre running, they have time to innovate.

During a panel discussion at the launch, Richard Fichera, vice president and senior research analyst at Forrester, alluded to the issue.

"Most businesses spend much more money on planned downtime than unplanned downtime... [but] unplanned downtime is a constant headache," he said.

"[However], don't make servers too smart as people like me may well be out of a job, as well as the IT people you sell to. You have to be careful of that."

When IT Pro asked Dave Donatelli, executive vice president and general manager of enterprise servers, storage and networking (ESSN) at HP, whether this increased automation would put IT admins out of work, he claimed it would actually create jobs.

"When it comes to IT budgets, 70 per cent of what they spend is running legacy applications," he said. "If you look at what businesses want to do, they don't want to spend 70 per cent of their IT dollars getting no new innovation, they want to spend that on innovation," Donatelli said.

"The biggest frustration facing IT [departments] today... is [they] aren't moving fast enough with [technology] innovations. So really what we have set out to do, be it with these servers or with converged infrastructure, is to allow them to spend more on innovation and less just keeping what they already have running.

"I think this is job positive and job creating, particularly if you look at the technology coming out, more jobs for innovating."

Mark Potter, senior vice president and general manager of industry standard servers (ISS) at HP, backed up his colleague and claimed IT admins were not concerned but excited about the technology.

"IT admins are celebrating," he told IT Pro. "They are celebrating that they don't have to spend all their time keeping their datacentre running, they have time to innovate. They applaud this [automation]."

For more information on the new ProLiant Gen8 server range launched by HP this week, click here.

The ProLiant Gen8 portfolio was born after two years of research and development, costing the company $300 million (191 million), titled 'Project Voyager.'

The research began because HP realised servers needed to "change course" to keep up with "the era of the cloud," claimed Mark Potter, vice president and general manager of Industry Standard Servers (ISS) at HP.

We need to free up admins, take advantage of capacity and remove costly unplanned downtime.

"Manual processes cost so much time that could be used to innovate," he said, during a keynote speech at HP's first global partner conference in Las Vegas.

"They don't allow the customer to harness the stranded capacity in a datacentre and address growing energy needs. We need to free up admins, take advantage of capacity and remove costly unplanned downtime."

As a result, the main focus of the ProLiant Gen8 servers is automation. An update to HP's Sea of Sensors technology will enable admins to see the best place to put a server, considering power, cooling and redundancy circuit issues through one tool. Potter claimed the server can then be plugged in and ready to go straight away.

It also automatically provisions power to certain areas, making sure nothing is wasted, and agentless software continues to cut the need for manual processes.

Potter said these examples - which fall under the umbrella of HP's 'ProActive Insight' architecture - made "a brand new ground breaking architecture, to build servers with automation and intelligence to take care of itself; exactly what it should do in the cloud era."

Other new features of the range include a smart socket addition, reducing the risk of bending pins when putting in a processor - a common cause of failures in the datacentre.

Smart Drive technology combines with the automation functionality to give engineers better information on which drives are where, making sure they don't remove the wrong one - another common cause of breakdowns in datacentres.

HP listed a number of claims around performance, including 10 per cent less energy consumption, six times faster I/O for storage and three times productivity for admins, but companies will have to wait until March to get their hands on the new servers when they become generally available.

The ProLiant Gen8 family includes the ProLiant ML Tower - for remote or branch offices - ProLiant DL rack-mounted servers, and ProLiant BL blade servers and ProLiant SL scalable system servers - suited for cloud systems.

Confirmation of the launch timescale came today from Dave Donatelli, executive vice president and gemeral manager of enterprise storage, servers and networking [ESSN] at HP.

The servers were born out of the 'Project Moonshot' research programme developed by HP and announced at the start of last November.

The servers will be a significant boost to British chip designer ARM, which has traditionally been overlooked in favour of Intel and, to a lesser extent, AMD x86 processors.

This will lead to all kinds of incredible advancements and change the market

Donatelli took to the stage at HP's first global partner event in Las Vegas today, telling 3,000 HP partners his company was proud of the "game changing" announcements it had made in the server space in the past 100 days.

"Instead of building out of x86, we are going to build servers out of ARM chips and Atom chips, those same chips from your smartphones," he said.

"This will lead to all kinds of incredible advancements and change the market."

Donatelli then confirmed the new servers would be shipping in "the first half of the calendar year."

The servers themselves will be aimed at datacentres looking to run more effectively, especially large enterprises and cloud providers.

Project Moonshot aims to help companies with large sprawls consolidate their data centres and, in turn, make them more energy efficient.

HP has not turned its back on x86-based servers, however, and today launched its latest range, the ProLiant Gen8 family. To read more about the new servers, which also claim to make data centres greener with a 10 per cent reduction in energy consumption, click here.

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.

It has been a long wait for the chips, which contain more cores than any other x86 chips on the market today, as the industry has known about them for a year now.

With more cores on board, AMD is hoping cloud-focused manufacturers will jump on the new chips, which the vendor believes are the "natural choice for heavy-threaded cloud computing environments."

The wait for the most anticipated new product and architecture for servers is over.

Designed on AMD's new Bulldozer architecture, the Opteron 6200 will appear in Acer, Cray, Dell, HP and IBM servers in the "coming days and weeks," the vendor said.

There are 10 different models of the Opteron 6200, ranging from four cores to 16, with the maximum clock speed hitting 3.3GHz. AMD's Turbo CORE Technology allows for an extra power boost of up to 500MHz.

AMD claimed the new chips offered up to 89 per cent greater performance than other server processors costing the same price.

"Our industry is at a new juncture; virtualisation has provided a new level of reliable consolidation and businesses are now looking to the cloud for even more agility and efficiency. We designed the new AMD Opteron processor for this precise moment," said Paul Struhsaker, corporate vice president and general manager for Commercial Business, AMD.

"The wait for the most anticipated new product and architecture for servers is over. Leading OEMs are now offering cloud, enterprise and HPC customers a full suite of solutions based on the industry's most comprehensive server processor portfolio, the new AMD Opteron family of processors which deliver an inspired balance of performance, scalability and efficiency."

The chip maker is looking to make a big impact on the x86 market, which Intel still dominates. Much emphasis has gone on the power savings of the Opteron models.

"When these energy efficient processors are used in servers, the server power savings can be dramatic as much as a 20 per cent reduction in server active idle power and an 18 per cent increase in server performance-per-watt," said Andy Parma, a product marketing manager for at AMD, in a blog post.

AMD also announced its new Opteron 4200 chips, claiming it is the "world's lowest power-per-core server processor built to deliver unparalleled efficiency for complex, technical workloads."

There are eight different models for the Opteron 4200 design, with a top clock speed of 3.3GHz. Customers can choose between six and eight-core models.

The new offerings will see HP advise users on data centre strategy, operations and continuity, which it claims will speed up the implementation of cloud by 40 per cent a figure it came to by comparing competitor services.

Legacy data centres need a critical transformation to prepare for cloud computing.

The strategy segment of the consulting claims to contribute to this speed increase by enabling customers to plan design and build at the same time, whilst operations focuses on cutting down lifetime data centre costs by up to 37 per cent.

The continuity piece of the puzzle is more of a future gazing service that hopes to predict when things will go wrong in the data centre, meaning the business is prepared in advance to tackle data centre issues head on.

"Legacy data centres need a critical transformation to prepare for cloud computing," read a statement from HP.

"They must be architected with shorter planning cycles to deliver services instantly, while optimising operations and improving reliability of those services. This requires better alignment, communications and planning across the organisation."

"HP services address data centre strategy, operations and continuity with the expertise, guidance and proven methodologies clients need to succeed."

In addition, HP has updated its CloudSystem platform for simplifying building and managing cloud environments.

With HP CloudSystem Financing, customers are offered a variety of plans to pay for the upgrades, making the transition less about capital expenditure as the cloud should be.

HP AllianceONE resource centre is a "quick-start toolkit" available to either independent software vendors or service providers to give them an easy way to showcase their technologies in the cloud.

"Clients want our help in speeding their transition to cloud computing with proven methodologies and solutions based on HP Converged Infrastructure so they can increase innovation in their markets and better serve their customers," said Dave Donatelli, executive vice president and general manager of enterprise servers, storage, networking and technology services at HP.

"Our new offerings clearly reflect the best of HP's skills and technology in driving this evolution for clients, while reducing complexity, costs and risk."

A quarter of them withheld voting for the founder's re-election, as Mr Dell received less support than any other executive re-elected to the board.

The central issue facing the chairman and CEO was over a US Securities and Exchange Commission (SEC) investigation into how Mr Dell and his firm failed to adequately disclose information to shareholders over payments from Intel. The chip maker effectively paid Dell to keep using its processors, rather than switch to AMD.

The payments from Intel made it seem as though Dell was meeting or surpassing results expectations, the SEC said. Without them, however, the company would not have done so once between May 2001 and January 2006, the regulatory body claimed.

Dell was told to pay $100 million as part of the settlement, whilst the company founder personally paid out $4 million. It was not a pretty end to a murky story.

The legal investigation was not Dell's only problem. Some were calling for fresh blood, for a chief who could take Dell into a new era of IT in the cloud and thinner clients, as the desktop market falls.

This all came just after three years of Michael Dell's return as CEO of the company he founded, having stepped down in 2004.

Like a nerdy kid hitting the gym in earnest, Dell has bulked up to become a full data centre solution company.

He has made some bold attempts to answer questions surrounding his ability to take Dell forward over the last year, however. He has re-envisioned his company, whilst keeping a firm grip on the client market, as others namely HP have chosen to move away.

Has the company-coined "new Dell" brought him back from the brink and secured his place as the leader of the business he created?

Welcome back

From the outside, it appears so. Dell has been a confident company of late.

Back in February, it posted some solid results, as its fourth quarter smashed Wall Street expectations. Many pointed to a Dell revival, as the company shifted into higher margin business, including a greater focus on servers, storage and services.

In August, the company again posted some decent figures as net income went up 63 per cent year-over-year. However, it reduced its revenue expectations for the year down to between one and five per cent. Yet it would be harsh to slam Dell for this mixed showing, given the entire market is being hit by reduced demand, added to the fact it is going through a major transitional period. We can forgive this faltering for now.

Highlighting its commitment to change, Dell has spent a penny or two. As part of its redefining of its business to focus on overall solutions, moving away from a hardware-centric model, Dell has bought a raft of companies.

The acquisition of Perot Systems for around $3.9 billion kicked off the spending spree in 2009, marking a notable moment in Dell's services commitment. Then came KACE Systems in 2010, providing hardware-based systems management appliances, followed by storage company Compellent, having missed out on 3PAR. Its latest move was to get hold of Force10 Networks to provide some prowess in the networking space. Like a nerdy kid hitting the gym in earnest, Dell has bulked up to become a full data centre solution company.

"Finally, some of the power brokers in Dell have accepted that pile them high, sell them cheap' has had its day, and the acquisitions Dell has been making are all aimed at moving it into a much more complex, full hardware stack sell, with the software coming from solid relationships, mainly with Microsoft, but also on the Linux side as well," said Clive Longbottom, founder of analyst firm Quocirca.

"I do believe he has now seen that Dell could not go on as it was. New blood has been brought in, new approaches are being brought to the fore, and Dell is looking stronger than it has done for a while."

Lingering issues

On the face of it, Mr Dell has helped create a new company ready for today's IT world. Yet significant stumbling blocks remain. Number one is the declining PC market and Dell's falling position within it.

During its inaugural Dell World conference in Austin last week, Lenovo became the number two PC seller, pushing the US firm down into third place. Given the PC industry is already in decline, the Texan firm will not want to fall from the upper echelons of the market.

Exacerbating things is Dell's performance in other client segments. In tablets, one of the causes of the PC market's contraction, Dell has failed to impress. Its lineup of Streak devices have performed poorly, with the Streak 5 model ditched earlier this year. It has plans for a global release of 10-inch consumer-focused tablets, but they have not emerged in the West yet.

Dell hasn't made much of a mark on the smartphone space either. Despite producing a quality phone with the Dell Venue Pro, its market share remains infinitesimal. According to Gartner, its not even in the top 10.

Michael Dell may have to face up to the reality it shouldn't bet on the consumer segment as much as it currently is. Cisco has proven how disastrous major gambles in the consumer sphere can be, especially if you don't have the relevant acumen or experience.

By turning his ship towards a cloudy horizon and building hardware for the entire stack, Dell is looking like a considerably more attractive proposition.

In tablets, Dell will have to create something with top-level design quality and some significant power, as well as a low-entry price point if it wants to have any chance at succeeding. Whilst it can certainly create quality hardware, offering low costs looks unlikely. For starters, it won't be able to rely on apps and software sales like Amazon can.

Furthermore, no matter how much Dell says smartphones are part of the end-to-end solution for businesses, it can't deny the devices are primarily targeted at consumers. When you think of smartphones though, you do not think of Dell.

Simply looking at the most recent results, which showed much of the positives were coming from the enterprise segment, it should be clear to Michael Dell that placing considerable emphasis on the consumer is a big risk.

Let's get positive

Yet Mr Dell has done well. By turning his ship towards a cloudy horizon and building hardware for the entire stack, Dell is looking like a considerably more attractive proposition for companies who want a full business package.

"As long as Dell retains focus and manages to keep fighting with its broadened arsenal of compute, storage, management, network hardware, brought through with professional services and being sold as a solution sell, rather than a portfolio sell, I believe that it will begin to take business away from a wounded HP, and could also chip away at IBM where some organisations see IBM as being too expensive and cumbersome to deal with," Longbottom told IT Pro.

Michael Dell's second reign as CEO won't be of Steve Jobs proportions. He will not turn Dell into a cult-inspiring, behemoth of a tech provider, but there is little doubt he has steered the business in the right direction. All he has to do now is keep firmly on course hopefully steering away from consumer clients - amidst turbulent seas.

As IT spending looks set to diminish over the next year, Dell will have to look to those markets it knows will provide it with solid income.

Blade servers may have been introduced over five years ago as high processing density alternatives to rack server but their other benefits are now making them look a more sensible choice where rising costs are a big issue. Businesses are now less concerned about processing density and more concerned about the number of servers within a particular power envelope for a given rack size. To cater for this blade server design has evolved over the past couple of years and the focus has been moving away from sheer processing power to other areas such as reduced power consumption and heat output.

Cooling has been made more efficient as the chassis features have been developed so fans can be removed from the server blades themselves thus freeing up more space for memory, storage and faster processors. This has added benefits as a rack full of blade servers will have fewer moving parts and the chassis can deliver high levels of cooling and power redundancy. Rising supports costs are another consideration and blade servers have the facilities to provide better centralised management and easier deployment.

In this group test we take a closer look at the four big players in the blade server market and bring you in-depth reviews of the very latest systems from Dell, Fujitsu Siemens, HP and IBM. Having followed blade server development since its early days it has been interesting to see that these four blue chips now have the UK and European market pretty much sewn up. Blade server development is a costly business and many other vendors have either tried their hand briefly or simply stayed well away. Bizarrely, NEC came in a couple of years ago with its Express5800 blade system but a design fault on it UK web site meant that the product isn't currently being made available to this market. Tiawanese manufacturer Supermicro has been threatening to deliver a blade server for years but nothing has ever materialised whilst others such as Tatung and Nexcom have an insignificant or non-existent presence in this neck of the woods.

As you'll see in this group test, the four participants haven't rested in their laurels and are continuing to develop blade server technology with some impressive results. The latest dual-core Xeon and Opteron processors provide significant benefits and all four systems offer support for both of these with a resultant increase in processing density. And you're not tied to one technology either as blades allow you to mix and match processing requirements as demand dictates. Storage options continue to improve and many blades support the latest low-profile SAS and SATA hard disks with RAID as a standard feature. Connectivity options have come a long way as well with basic Gigabit Ethernet switching now complimented with 4Gbps fibre channel and in some cases even 10-Gigabit options.

Businesses faced with spiralling costs of managing and maintaining data centres and hosted services need to look closer at blade servers as they can be a better alternative to rack servers. Even with a relatively small number of vendors in this market there's still a huge range of choices available and we give you the low-down on the top four systems currently available.

Dell entered the blade server market way back in 2002 with its in-house developed PowerEdge 1655 product which was found lacking in a number of areas and was subsequently quietly withdrawn a couple of years later. Its next move came with the 1855MC which was the result of a collaboration with Fujitsu Siemens as Dell sourced from the same manufacturer. This arrangement continues with the 1955 so Dell's offering is very similar in many ways to the Fujitsu Siemens BX600 S2.

There are no differences with the chassis so you get the same number of blade slots plus power and cooling arrangements. Four expansion slots are provided at the rear and Dell also offers an extensive range of modules. The PowerConnect 5316M switch blade provides sixteen Gigabit ports with ten internal ports linked to the chassis backplane and six presented to the outside world. Two of these allow the blade's dual Gigabit ports to be routed through to different switch modules for redundancy. Two more bays are provided for another pair of PowerConnect 5316M switch blades but to use these you'll need to install an extra dual-port Gigabit Ethernet daughtercard on each server blade.

With Dell's QLogic FC daughtercard installed in the blades you can add Brocade Silkworm 3014 or McDATA 4314 FC switches or a simple FC pass-through module. Dell also offers a TopSpin Infiniband card as well.

The latest blade offerings are far fewer than those available for the BX600 as Dell has only one new server module that supports Series 5000 and 5100 processors. There's room for a couple of hot-swap 2.5in. SAS hard disks at the front and the server has an embedded SAS controller which provides RAID-1 mirroring. No other storage expansion options are currently available as Dell does not offer the same optional SAS and SATA extension modules as Fujitsu Siemens. Its fervent commitment to Intel has also boxed it into a corner as it currently has no Opteron blades whereas HP, IBM and Fujitsu Siemens all offer extensive support with the later capable of delivering a remarkable 16-way Opteron blade.

Installation of the 1955 is fairly straightforward which is just as well as Dell's documentation is truly a mess as you get one generic disk covering every product it's ever made. Consequently, we found it a real pain having to constantly swap through a multitude of manuals when configuring the chassis, the KVM module, the remote management controller, the switch modules and the server blades. As you'd expect, the chassis comes with the same advanced KVM module as the BX600 which provides KVM over IP and a connector for locally attaching a monitor, mouse and keyboard.

Dell's DRAC/MC remote management controller is physically the same as that fitted in the BX600 so you get serial and networks ports for local and remote chassis management. A quick visit to the serial port and CLI is required to assign the module an IP address and then you can move over to its web interface. Note that Dell's base specification comes with one management controller whereas Fujitsu Siemens only offers the BX600 S2 with the extra redundant module as standard.

It may be the same hardware but the web interface for the DRAC/MC is completely different to that offered by Fujitsu Siemens and in our view is not as well designed or as accessible. The home page opens with a chassis summary showing the module and logged in users along with a table displaying the status of each installed module such as power supplies and fans. A sensor log provides operational details on these components and the power budget section keeps you posted on the power draw of the chassis and each server blade.

Power for the chassis and switch blades can be cycled from this interface and you can access individual blades and power them up and down as well. Full remote control sessions allow the server blade and OS to be accessed and virtual floppy and CD-ROM support means it can be booted from an image stored on another system. Power management isn't up to the high standards set by IBM and HP as all you can do is create chassis redundancy policies. These keep power capacity on standby to avoid a chassis failure should a power supply fail. However, there are no tools to directly access the processors for controlling power and heat output.

For general remote access to the server blades Dell bundles its standard OpenManage Suite - the same as supplied with all its PowerEdge servers. Each server blade can be locally and remotely managed using the Server Administrator which offers browser based access and provides plenty of information about motherboard sensors. All operational data is presented in tabular format and you can see the status of the processors, power, hard disks, temperatures and voltages. Alert management allows faults to be linked to actions such as network broadcasts, sending e-mails and running an application.

The IT Assistant targets Dell-centric networks and can manage any of its servers, workstations and laptops running the appropriate agent software. It requires MSDE2000 which is included on the product CD-ROM and uses this to store basic inventory information about each system. An auto-discovery routine scans the network, creates a list of all discovered systems and then polls them regularly to make sure they are available. It provides a comparatively basic range of blade management tools which includes inventory and remote BIOS updates plus you can remotely monitor critical components and link problems with alerts.

If you want full power provisioning and management then you need to look elsewhere as apart from the basic power budgeting available from the DRAC/MC card the 1955 doesn't let you manage blade power consumption to same degree that HP and IBM allows. Software deployment is also conspicuous by its absence with no tools provided for automated installation of operating systems to multiple blades.

There's no denying that if Dell wants to keep up with the competition in the server market it has to have a blade server portfolio. The latest 1955 definitely represents good value but the price you pay is a more limited level of features and expansion options, a smaller server blade choice that doesn't include AMD and more basic management, monitoring and provisioning tools.

Verdict

A good value blade server package with plenty of connectivity options but Dell offers a very limited choice of blades and processing options whilst management tools are not designed specifically to handle this type of hardware

Chassis: 7U enclosure: ten blade slots; up to four hot-swap 2100W power supplies (two as standard); 2 x hot-plug fan modules; advanced KVM switch module with KVM and 10/100BaseTX ports; one hot-plug DRAC/MC management module with 10/100BaseTX and 9-pin serial port ; four interconnect bays - from £2,113 (exc. VAT) Interconnect modules: Cisco 3030 Gigabit - £2,239 (exc. VAT) PowerConnect 5316 Gigabit - £370 (exc. VAT) McData 4314 2Gbps FC - £4400 (exc. VAT) Brocade 4016 2Gbps FC - £4389 (exc. VAT) McData 4416 4Gbps FC - £4400 (exc. VAT) Server Blades: PE1955 - 2 x Xeon 5150 2.66GHz; 2GB 667MHz FB DIMM; Intel E5000P chipset; 2 x 73GB 10K SAS hot-swap hard disks; Dell SAS/5ir SAS chipset with RAID1; 2 x Gigabit Ethernet - £3,059 (exc. VAT)

HP was first to deliver in the UK blade server market with its compact BL e-Class, which set a very high standard in terms of features and processing density. Since then it's gone from strength to strength and the latest Blade System c-Class represents a completely new architecture that has been designed from the ground up.

HP has turned to its SuperDome family and taken concepts from these to improve performance whilst the Himalaya non-stop servers provide inspiration for fault tolerance. Management capabilities come from the ProLiant server products along with their iLO (integrated lights out) features and the c-Class is designed to integrate with all of HP's standard management software.

At its foundation is the 7000c enclosure which stands at 10U high and offers sixteen half-height or eight full-height server blade slots. It has a 5Tb/sec mid-plane which also incorporates design concepts from Tandem systems. All sixteen slot connectors are routed through to multiple power supplies and network expansion modules and each blade has its ports directed to different expansion slots for network connections. Power is particularly well catered for as the chassis has six power inputs at the rear each fed by a hot-plug supply. Fault tolerance is extremely good as you would need to lose four power supplies before a server blade slot would fail.

The new cooling modules are rather smart as they are based on jet engine design principals with the aim of achieving maximum efficiency and low noise levels. A quick peek down the end of a fan module is enough to confirm this as the metal fan is, indeed, shaped more like a turbine. All air flow from the fan modules is spread across the entire chassis so each blade slot does not have any dedicated modules. Louvres in each blade slot only open when a fan module is present and thermal sensors in the chassis can control fan speeds dynamically. If you do need to replace a fan module you'll be pleased to know they only cost around 80 each.

An operator panel and LCD screen is provided on the chassis for basic system commissioning, fault analysis, assigning IP addresses to individual blades and checking on general system health. Usefully, it provides a chat mode where text based conversations can be conducted between a remote manager and support staff local to system. Up to two management blades can be installed for redundancy and each is equipped with network and serial ports plus an embedded iLO 2 chip. Connectivity options abound as the chassis has eight slots at the rear and HP offers an extensive range of modules. These include 16-port pass-through blades plus Gigabit switches which have been designed by Cisco Systems specifically for HP. Other modules include 12 and 24-port Brocade FC SAN switches and another with eight copper Gigabit and four mini-GBIC slots with SFP modules available for SX, LX and LH fibre connections.

HP's virtual connection module tackles the problems of blade deployment in large businesses with complex security and change management procedures. It allows you to use your own MAC address pool so when a blade is replaced with a new one it can be assigned an address from a virtual pool. This is designed to speed up deployments and upgrades and is quite secure as HP creates and registers the MAC addresses in the pool for you. A virtual connect module with a pool of WWNs for FC SAN deployment is expected later this year.

HP offers an extensive range of server blades with options for half-height and full-height models. It is focusing heavily on Intel's Series 5100 processors and although it can make the 5000 processors available it only sees these as suitable for validation purposes only. HP has also standardised on SAS storage across the range as it provides the best performance. It doesn't recommend using SATA hard disks with its blades but you can install them if you wish. Build quality and design is very good with the DL460c half-height blade, for example, offering two hot-swap drive bays. The processors are mounted underneath and topped off with solid passive copper heatsinks. The storage backplane comes with an embedded Smart Array RAID controller and cache memory which can be upgraded with an optional 128MB cache module with battery backup pack. Further back are two expansion slots which are based on PCI Express technology and accept Gigabit NIC and FC cards.

Network options are particularly good as the blade has a pair of embedded Gigabit NICs which incorporates a TOE and hardware iSCSI initiators. The full height blades deliver greater storage potential with room for up to four SAS and SATA 2.5in hard disks and the storage controller delivers support for RAID0, 1, 5 and 6. Three free slots are available and support extra Gigabit and FC expansion cards. Remote management is on the cards for all HP's blades as each one also has an embedded iLO 2 chip on board. HP goes with the flow to AMD as well as it also plans to offer a comprehensive range of Opteron based server blades.

HP's Onboard Administrator provides full remote browser access to the chassis and its home page opens with a system status readout showing colour coded icons for errors and faults. A wizard is also provided to help with initial installation. An Explorer style menu makes light work of accessing the various functions and photo-realistic pictures of the chassis provide an at-a-glance view of installed components. Power control is very high on the agenda as the Onboard Administrator provides an instant readout on chassis power consumption, power available, current BTU/hr, max BTU/hr and temperatures. The total VA power draw for the entire chassis can be set from this interface and a dynamic power saving mode will attempt to run some supplies at a higher rate so that it can put others into a standby mode. Power controls extend to the blades where the processors can be controlled dynamically to limit power consumption. We also found the blade iLO management interface as easy to use as that offered by standard ProLiant servers with good access to remote control, power and diagnostics options.

HP's Insight Control Center Edition provides more in-depth management functions as the basic product includes the Systems Insight Manager tools which include enhanced browser-based remote management and monitoring. It offers high levels of information about general operations and extensive alerting facilities and can remotely access any HP blade with an Insight agent installed. You get tools to help deploy software to multiple blades and HP can supply pre-prepared disk images of a variety of operating systems. There's even more as you get tools for vulnerability and patch management, BIOS and driver version control and performance monitoring and alerting.

The Blade System c-Class is without doubt one very classy blade server with a range of sophisticated features and management facilities. One drawback is the lack of backward compatibility with previous generations of HP's blade servers but if you're new to the game then the c-Class has everything needed to run critical enterprise applications with mainframe level fault tolerance.

Verdict

A superbly designed system with masses of fault tolerance, plenty of server options and quality management features - blade server technology simply doesn't get any better than this.

Chassis: 10U rack enclosure; 8 full height/16 half height slots; up to six 2250w power supplies (min. two supplies), up to ten hot-swap cooling fans (min. four fans); up to two management modules with 10/100BaseTX, serial and embedded iLO 9one as standard); eight interconnect slots; operator panel and LCD display - from £2,421 (exc. VAT) Interconnect modules: 16-port Ethernet Blade Switch - £737 (exc. VAT) 16-port Cisco Catalyst 3020 Blade Switch - £2,527 (exc. VAT) 12-port Brocade 4Gbps FC SAN switch - from £3,159 (exc. VAT) BL460c server blade: half-height chassis; 1 x 1.6GHz Intel 5110 Xeon; Intel 5000P chipset; 1GB 533MHz FB DIMM memory; 8 DIMM slots; 2 x hot-swap SAS/SATA hard disk bays (no disks); Smart Array E200i Controller with 64MB cache; 2 x Gigabit Ethernet with TOE and iSCSI; 2 x expansion slots - from £1,065 (exc. VAT) BL480c server blade: full height chassis; 1 x 1.6GHz Intel 5110 Xeon; Intel 5000P chipset; 2GB 533MHz B DIMM memory; 12 DIMM slots; 4 x hot-swap SAS/SATA hard disk bays (no disks); Smart Array P400i Controller with 256MB cache; 4 x Gigabit Ethernet with TOE and iSCSI; 3 x expansion slots - from £1,545 (exc. VAT)

Fujitsu Siemens has been in the blade server business almost as long as HP, delivering its first system - the BX300 - only a couple of months after HP unveiled its first BL e-Class back in 2002. The BX300 was replaced by the BX600 in 2004 which delivered support for Xeon processors and this latest S2 generation brings in a number of interesting features that make it quite unique.

Standing at 7U high, the BX600 chassis hasn't changed much over the years and still has room for up to ten server blades. Four hot-swap redundant power supplies are still the order of the day but these have been beefed up from the 1200W of its predecessor to 2100W to cope with the new processing power requirements.

The chassis has a single mid-plane which all the blades connect to for receiving power and network connectivity and all redundancy in these areas is determined by the modules fitted at the rear. The two upper slots are provided for Ethernet switch blades with the standard modules each delivering ten internal Gigabit downlinks to the server blades via the mid-plane and six external Gigabit network uplinks. They support remote web browser management and the two network adapters on each server blade are routed through to different switch modules. There are plenty of other switch blade options and Fujitsu also now offers one which provides six copper Gigabit ports along with Infiniband and 10GBase-CX4 copper 10GbE connections.

The lower switch blade slots have a dual purpose as they can be populated with FC blades manufactured by Brocade with each offering six 4Gbps FC ports. However, the server blades have options to fit either FC or Gigabit daughtercards and if you go for the latter then the lower chassis slots need to be populated with extra Ethernet switch blades instead. Fujitsu Siemens only offers the chassis populated with both hot-swap management modules - Dell will sell you a system with only one. Each module is equipped with a serial port for CLI access and a 10/100BaseTX port for remote browser-based control. A small module alongside one of management blades provides a separate KVM switch for local blade access.

Fujitsu Siemens offers plenty of Intel based blades with support for the latest Series 5000 and 5100 dual-core Xeon processors. However, its support for AMD blades has allowed it to deliver a multiprocessing solution to market that makes the BX600 currently quite unique.

You can populate the chassis with blades using the dual-core 200 Series Opteron DP processors but opting for blades with the 800 Series MP Opterons brings some very interesting scalability options into play. Take a pair of these Opteron MP blades and you can bolt them together and using the slots in the casings insert two bridge boards that link the blade's HyperTransport buses together. Another side band connector links the blades system boards together so all I/O can be routed through them both.

It get even more interesting as you can take two of these dual-blade modules and connect them together to create a single, 16-core Opteron processing blade. Storage is a casualty as the lower drive bays needs to be removed as the two dual-blade modules are joined using a special four-port side band connector.

However, you're still left with four drives so RAID-10 or RAID-5 with hot-standby is a distinct possibility. Fujitsu Siemens is quite rightly very proud of these capabilities and it is a feature that differentiates the BX600 clearly as Dell does not support this feature and has no AMD blades at all. Storage options are up with the rest as the blades can support SAS or SATA hard disks in 3.5in. and 2.5in form factors. The front of each blade provides plenty of status information along with power and KVM selector buttons and a proprietary port allows USB devices and a monitor to be connected with the supplied Y-cable.

Server and chassis management tools are in abundance as both the management and switch blades can be monitored and configured from their respective web interfaces. The former offers very good access and opens with a complete status report on all installed components and environmental values such as fan speeds and temperature. An Explorer style menu tree to the side allows you to drill down and view individual components. With power supplied to the chassis the management blades provide light-out access so you can power the entire system up and down remotely.

Individual blades can be accessed from this interface where power can be controlled manually or to a custom schedule on specific days. The new Ethernet switch blades offered with the BX600 see some major improvements over the older three-port modules as not only are they Layer2/3/4 routing switches but they are designed to integrate more easily into Cisco-centric networks by supporting CDP (Cisco Discovery Protocol).

For general chassis and blade management you get the ServerView S2 Java utility which is now the standard tool for all PRIMERGY servers - the old ServerView Windows utility is no longer being developed. This S2 utility accesses any system running the relevant agent component over SNMP and provides a fine array of operational information. You can monitor areas including processors, temperatures, voltages and fans and tie in alarms or errors to notifications by email, network messages or pagers. The power supplies can also be remotely accessed so you can schedule server power cycles and use a watchdog timer to automatically reboot a server blade if its hangs. The ASR (automatic server restart) tool allows you to set thresholds on temperatures and force a server shutdown if these reach critical values.

For OS management, Fujitsu Siemens wrote its own deployment and disk imaging software and bolted this onto its standard ServerView management software. You start by installing an OS on the first blade and then add the extra Windows utilities to turn this into a deployment server. You then install an OS on the next server blade in line, take an image and store it on the deployment server. Next, you place the target blades in a group and fire up the RemoteDeploy process which multicasts the selected image to each server simultaneously.

It may come from the same stable as Dell's PowerEdge 1955 but the BX600 S2 delivers a far superior choice of server blades and more storage options. It may not be as well featured or as manageable as the c-Class and BladeCenter H systems but it does make up for this with good overall value and some very interesting blade expansion options.

Verdict

Not as sophisticated as the systems from HP and IBM but the BX600 does deliver excellent build quality, a good range of storage options and some unique Opteron based blades

Chassis: 7U enclosure: ten blade slots; up to four hot-swap 2100W power supplies (two as standard); 2 x hot-plug fan modules; advanced KVM switch port with KVM and 10/100BaseTX ports; two hot-plug management blades with 10/100BaseTX and 9-pin serial port ; four interconnect bays - from £2,826 (exc. VAT) Interconnect modules: Gigabit Ethernet pass-through - £546 (exc. VAT) Gigabit Ethernet 6-port - £908 (exc. VAT) Brocade six-port 4Gbps FC - £5,624 (exc. VAT) Server Blades: BX620 S3 - Xeon 5050 3.0GHz; 1GB 533MHz FB DIMM; Intel E5000P chipset; 2 x SAS/SATA hot-swap hard disk bays (no disks); LSI Logic SAS chipset with RAID1; 2 x Gigabit Ethernet - £1,334 (exc. VAT) BX630 - 2 x AMD Opteron 246 2.0GHz; 1GB 400MHz SDRAM; AMD 8131/8111 chipset; 2 x hot-swap SAS/SATA hard disk bays (no disks); LSI Logic SAS controller with RAID1/Promise SATA controller with RAID0 and 1; 2 x Gigabit Ethernet - £1,605 (exc. VAT) BX630 Quad - 4 x AMD Opteron 865 1.8GHz; 1GB 400MHz SDRAM; AMD Thor chipset; 4 x SAS hot-swap hard disk bays (no disks); LSI Logic SAS chipset with RAID1; 2 x Gigabit Ethernet - £5,423 (exc. VAT)

IBM has always had a very strong presence in the blade server market since its onset and now offers the largest choice of chassis with three different models available. The BladeCenter H on review is the very latest to join the family and is aimed at a wide variety of functions including running mission critical applications.

Slightly smaller than HP's c-Class, the BladeCenter H chassis comes in at 9U high and has fourteen full height blade expansion slots. A key feature of this chassis is internal design as the chassis uses a mid-plane which is physically split in two with regard to both power and networks connections. The server blades have two, separate connectors which mate with both halves of the mid-plane and this arrangement allows the chassis to provide power and network redundancy to each individual blade.

When two switch modules are installed, the Gigabit Ethernet ports on each blade are routed to each one. One feature that makes IBM's blade server offerings stand out as unique is they are all designed to be backward compatible with all previous chassis. This allows blades from older systems to be used in newer chassis but, more importantly, also allows older blades to be replaced with newer high performance modules.

At the rear, the centre of the chassis is taken up by two enormous blower fans. With a totally populated chassis the maximum power draw on the chassis is in the region of 3.6KW so cooling needs to be good. If one blower module fails the other goes into a turbo mode to ensure cooling isn't reduced and is capable of shifting some serious volumes of air. IBM advised us that the cooling system on the BladeCenter H is actually over specified to the point where it can handle a total power draw of 8KVa.

Expansion slots are arranged on each side of the fan bays. The outer slots accept a wide variety of networking modules and the upper and lower ones mate with different parts of the mid-plane. The chassis accepts up to two management modules which provide a local monitor and two USB ports for connecting a keyboard and screen.

This arrangement provides local KVM facilities but the Fast Ethernet port on each module also delivers KVM over IP so the chassis and blades can be remotely managed simply by pointing a web browser at the management module's IP address. There are plenty of network connection options with Gigabit Ethernet modules developed by Cisco and Nortel, a choice of Brocade, McDATA or QLogic 2Gbps and 4Gbps FC SAN blades and copper pass-through blades. Two bays at the top and bottom of chassis accept a new module which has been co-developed between IBM and Cisco. These are unique to the H chassis and provide high-speed Infiniband connections and 10GbE to those blades that have the appropriate mezzanine card installed.

The management modules provide direct browser access to the chassis with the home page offering a general overview of the chassis and installed blades. It's not as sophisticated as HP's c-Class chassis interface but it provides equally good levels of information and access. You can keep an eye on event logs and environmental values while a fuel gauge tells you all about power consumption, the total available and whatever is left in reserve. It provides full access to each blade and allows you to remotely control them, cycle power and update firmware.

Configuration options are as good and we liked the feature that allows blade boot sequences to be changed on the fly without requiring a reboot. The switch modules all have their own web management interface and controls can be implemented to ensure that only specific users can access them.

Director is the foundation of IBM's server management portfolio and is the standard tool for managing IBM servers, workstations, laptops and, of course, its blade servers. Even without agents installed, Director will pick up the blade server chassis and discover all the installed components including switch modules. However, the agent provides benefits such as enhanced inventory details plus support for remote software distribution and firmware upgrades. A range of tasks including inventory, blade management and hardware status are displayed in the right hand panel and can be activated by dragging and dropping them onto the relevant system displayed in the centre pane.

Operating system distribution is handled by the Remote Deployment Manager (RDM) package which is IBM's standard tool for all its servers, desktops and laptops. However, it does also resell Altiris' deployment software as an option. RDM is a plug in to Director and provides all the OS deployment facilities. Systems can be booted using PXE and RDM uses profiles to store global settings for each OS. It can use either Symantec .PQI disk image files or its own CloneIt files to install an OS on target systems. Linux, Windows and VMware ESX deployments are all supported and you can schedule deployments of your chosen OS to multiple blades.

IBM's Power Executive plug-in is available as a free download and provides real time power monitoring of entire BladeCenter chassis and individual blades and each pair of power supplies can be placed within separate power domains. The current version only supports real time monitoring but new firmware upgrades in the future will provide power capping which allows CPU clock speeds to be modified to reduce power consumption.

As a governing member of Blade.org, IBM has embraced an open design which allows different manufacturers to create modules and switches for the BladeCenter. Its server blade offerings are equally impressive as it has versions that support Intel's Xeon DP, Series 5000, 5100 and Xeon LV processors. Blades are available with AMD single and dual-core Opterons and IBM uniquely supports PowerPC as well. It even offers cell processor blades from Mercury Computer Systems which are designed for applications such as data mining.

The blades themselves are very well built and the LS21 dual socket models can be upgraded to four processor sockets by removing lid and fitting a second blade on top. All have IBM's light path diagnostics onboard and a capacitor allows the blade to be removed with the lights still activated so you can see what the fault is. Optional blade storage expansion modules are also available which offer up to three hot-swap 2.5in. SAS drives allowing RAID-5 to be extended to each blade at the cost of an extra blade slot.

A key feature of the BladeCenter H that makes it stand out is choice as IBM offers four vendors for processors, two different deployment tools, Cisco, Nortel and IBM Ethernet switches plus three SAN switch vendors. Build quality and expansion are exemplary and although its Director management software is starting to look long in the tooth it does provide some sophisticated power controls. The company is also currently promoting its backward compatibility capabilities with some tempting incentives.

Verdict

A superbly built and well specified blade server with plenty of expansion options plus a massive choice of blade processors and connectivity options. Its Director software could do with a refresh but it does deliver good access to the blades along with quality power management tools

Chassis: 9U rack enclosure; 14 server blade slots; up to four hot-swap 2900W power supplies (two as standard); two blower fans; up to two advanced management modules (one as standard) with 10/100BaseTX and KVM port; ten interconnect slots; media tray with DVD-ROM, 2 x USB 2.0 and light path diagnostics - from £3,568 (exc. VAT) Interconnect modules: Server Connectivity Module; £770 (exc. VAT) QLogic(R) 10-Port 4Gbps FC Switch Module - £5,815 (exc. VAT) Server Blades: HS20 - Xeon 2.8GHz; 1GB 400MHz SDRAM; E7520 chipset; 2 x Ultra320 hot-swap hard disk bays (no disks); LSI Logic Ultra320 chipset with RAID1 and 1E; 2 x Gigabit Ethernet - £1,112 (exc. VAT) HS21 - Xeon 5110 1.60GHz; 1GB 667MHz FB-DIMM; Intel 5000P chipset; 2 x hot-swap SAS hard disk bays (no disks); LSI Logic SAS chipset with RAID0 and 1; 2 x Gigabit Ethernet - £1,472 (exc. VAT) LS20 - AMD Opteron 250 2.4GHz; 1GB 266MHz SDRAM; AMD 8131/8111 chipset; 2 x Ultra320 non hot-swap hard disk bays (no disks); LSI Logic Ultra320 chipset with RAID1; 2 x Gigabit Ethernet - £1,584 JS21 - 2 x PowerPC 970 2.7GHz, 1GB 533MHz SDRAM; IBM U4 chipset; 2 x SAS non hot-swap hard disk bays (no disks); IBM SAS controller with RAID0 and 1; 2 x Gigabit Ethernet - £1,897 (exc. VAT)

Blade systems represent one of the fastest growing sectors in the global server market, but most are way beyond the financial means of SMBs. HP spotted this gap a while ago and launched its BladeSystem c3000 Shorty' to fill it. Fujitsu now follows suit with its new Primergy BX400 S1.

The BX400 S1 targets SMBs and remote offices that want to amalgamate all their IT services in a single, low maintenance system that's easy to deploy and manage. It has plenty of expansion potential and is designed to work in normal office environments without the need for extra air conditioning or cooling.

Physically, the BX400 S1 is very similar to the c3000 - both are 6U high and built like tanks. The eight half-width server blade slots are accessed from the front and are arranged horizontally. The c3000 has vertical slots and accepts four full-height, eight half-height blades or a mixture of both.

The colour LCD display and control pad on the top panel are used to view the health status of the entire system and fire up a base installation wizard. This runs through chassis configuration, enabling management access and choosing a power management scheme.

The BX400 S1 supports up to four 1600W hot-plug supplies and each supply has an integral cooling fan. Noise levels are extremely low making so it's ideally suited to office deployments.

Server blade choices aren't as broad as HP's. Fujitsu has three Xeon-based server blades, whereas HP also has a number of AMD server options. The BX920 blade has two hot-swap SFF SAS/SATA disk bays at the front. To accommodate them the memory slot count has been reduced to nine so this model supports up to 144GB of DDR3 memory.

The BX922 has two internal cold-swap SFF SATA disks and twelve DIMM slots supporting up to 192GB of RAM. The BX924 blade supports up to 288GB of memory, but to make room for the eighteen DIMM slots it has two SATA SSDs fitted in cut-outs in the processor heatsinks.

This blade is aimed at virtualisation duties and supports remote booting or loading a hypervisor from an internal USB flash module. The other two blades have embedded four-port Gigabit Ethernet controllers, but the BX924 has two 10-Gigabit Ethernet ports.

All server blades have a riser tray at their rear that accepts a pair of mezzanine cards. Fujitsu currently offers Gigabit, 10-Gigabit, Brocade Fibre Channel and InfiniBand cards.

For storage blades you have the SX940 four-bay hot-swap model which occupies a single slot and uses a direct PCI Express connection via the chassis mid-plane to the server blade below it. The new SX960 is a double-height ten disk version, while for local backup there's the SX910 which has an LTO tape drive.

The soon to be launched VSX960 blade provides storage virtualisation services and comprises BX920 control and SX960 storage blades. The BX920 runs NetApp's Data ONTAP-v software in a VMware ESX virtual machine and presents the SX960 storage resources as NAS shares and iSCSI targets.

Four interconnect bays are available and Fujitsu offers a good range of connection options including Gigabit Ethernet, 10-Gigabit Ethernet, InfiniBand and 8Gbps FC blades. The chassis mid-plane is split into three fabrics with the first connecting to the embedded network ports on each blade. Fabrics 2 and 3 link directly to the server blade mezzanine cards.

Power consumption is very low and we measured the chassis with two supplies drawing only 136W. With BX920, BX922, BX924 servers and one SX960 storage blade powered up, total consumption only rose to around 520W.

The base system comes with one management blade and you can add a second for redundancy. These have dedicated network ports and provide web browser access for managing and monitoring everything in the chassis.

Remote management is easily the match of HP as Fujitsu's ServerView Dashboard provides direct web access to the chassis, servers, interconnect modules and power supplies. Graphics of the front and rear show which components are installed and each one can be accessed individually for management and monitoring.

The dashboard provides a readout on power consumption for the entire chassis, as well as each server blade. You can choose from six power schemes to strike a balance between performance and consumption. Server blade power is also controlled from the dashboard. Once switched on, the iRMC2 embedded controller gives direct web access for remote control.

One feature that speeds up OS deployment immensely is the ability to boot each blade in sequence from a USB stick in the shared I/O module at the top of the chassis. This loads Fujitsu's ServerStart OS installation tool and once the first blade has finished with the media module and is in the process of loading the OS, it will be passed on to the next blade and the process started automatically.

The new BX400 S1 is a close match to HP's c3000 as they are both similarly priced and equally well built. HP does offer a greater number of server blade options, but the low noise levels of the BX400 S1 make it a very good choice for office use.

The only other blade server that competes in this market sector is IBM's BladeCenter S. Its 7U chassis has greater expansion potential, but it correspondingly takes up a lot more space. It's also much noisier so if you want this in your office you'll need IBM's massive office enablement kit.

Verdict

It's taken a long time for Fujitsu to respond to HP’s c3000 but the wait looks worthwhile. Although Fujitsu's server blade choices aren’t as extensive as HP’s, the BX400 S1 is equally well built, just as easily managed and its very low operational noise levels makes it highly suited to life in the office.

Chassis: 6U enclosure with eight half-width blade slots Power: 2 x 1600W hot-plug supplies (max. 4) Fans: 2 x hot-swap (max. 3) Expansion: Four interconnect I/O module bays Management: Fujitsu ServerView Dashboard, ServerView Suite software Other: DVD+/-RW +/-DL, USB, Serial I/O module

At its annual Tech Forum held in Las Vegas, the company's executive vice president and general manager of Enterprise, told a press conference it was the "most significant" refresh in four years.

There are three new additions to the Proliant G7 server rack range, the DL980, DL580 and the DL585, all aimed at aiding HP's move to a converged infrastructure.

The products are built from what the company deems "industry standard" components, meaning whilst they all look very similar, upgrades, new parts and even power supplies are also easily interchangeable.

The DL980 is an eight socket server with 2TB of built in RAM, which HP claims doubles the performance of other x86 technologies already on the market.

The DL580 and DL585 are both four socket servers. The first offers up to 1TB of memory and 11 IO slots, with the claim it offers the same performance of 91 G4 servers in just one box.

All three servers feature new self healing capabilities, enabling the isolation of individual memory failures therefore increasing uptime, and come with a bold claim of 96 per cent reduction in power and cooling needs.

"The brand new Proliant G7 [servers] really are breakthrough," said Mark Potter, vice president of industry standard servers at HP, during the press conference.

"They have double the availability of competing technology and the ability to truly take advantage of the scale."

There will also be seven new additions to HP's leading Blade product line, including what the company claimed was the first ever 1TB memory server blade, the BL680.

One rack of the four socket blade has enough RAM up to 16TB to house the world's largest library, according to Potter.

On the other side was the BL620 which HP said was the most scalable two socket server "on the planet." It has fibre channel, iSCSI, 1GB and 10GB ethernet connectivity, helping again with HP's drive towards convergence.

All the new servers will be available by the end of July.

Forrest Norrod, general manager of Dell's server platforms, started off by praising both Intel and AMD, saying their latest processors offered a whole new level of capability.

The first product, the PowerEdge M710HD, was introduced as a no compromise half-height blade server. Despite its size it contains two Intel Westmere EP processors, each with six cores and up to 18 DIMMS sockets with hot swap SAS drives.

A Dell spokesman told IT PRO that due to thermal restrictions, competitors such as HP were only able to offer SSDs in the same size blade.

A key feature of the new launch is the redundant embedded hypervisor - in the form of ESX hypervisor - located on a redundant SSD card, which was first introduced into Dell products with the PowerEdge R910 last month.

Another unique feature is the use of male connectors on the mid-plane connector. It means in the event of a bent pin, the whole chassis would not need taking down to repair. The PowerEdge M710HD will be available in early July for 1,489 ex VAT.

The second product launched is the PowerEdge R715. It is an AMD based 2U system, offering 24 cores and designed for virtualisation or server consolidation. The R715 will be available from July 2010 for 1,659 ex VAT.

The third product was highlighted by Norrod as unusual. The PowerEdge M610x is a full height blade which offers two full bandwidth Gen2 16x PCI-E slots, providing 250W each. This enables customers to fit such add-in cards as GPGPU cards, such as Nvidia's Fermi.

"Our HPC customers are just giddy [about this]," declared Norrod, "[and] our online transaction customers are extremely excited about this," he added.

The M610x offers further delights by enabling the fitting of up to a terabyte of Fusion IO Flash, delivering as much as 120,000 IOPS. The M610x is available in early July for 1,659 ex VAT.

Finally came the M1000e chassis. It has been enhanced with an improved power supply and more efficient fans to boost power conservation.

Norrod announced enhancements to Dell's system management capabilities, with updates to its LifeCycle Controller and Chassis Management Controller. He claimed the former was unique in the market for automatically provisioning replacement components with the correct BIOS and drivers, saving administrators a great deal of time.

The Chassis Management Controller is also designed to give time back to administrators with a streamlined graphical user interface.

Availability for the M1000e chassis LifeCycle Controller and Chassis Management Controller was not confirmed at the launch.

With new rack, tower, multi-node and blade servers using the AMD Opteron 6000 series, Acer is hoping to take on the likes of Dell and HP.

At a press conference held today in Milan, Gianluca Degliesposti, vice president of global server business development at Acer, said: "[Acer] is like a car coming on the highway behind the likes of HP, IBM and Dell. We need to understand them, follow them and understand what they are doing but we also need to take over."

The first of the bundle of releases are two blade servers called the Gateway GB2x285 and the Gateway GB685. Both are aimed at high performance computing and enterprises, with the GB2x285 offering two dual socket nodes in one blade and the GB685 providing four separate CPUs in just one node.

The rack servers come with the model names GR385 and GR585. The first of the two offers up to 24 memory DIMMS, two 12 core CPUs and room for 16 HDDs. The second takes the title of the highest performer in the rack range with space for four CPUs in two rack units, which Acer claims will make it more cost effective to buy a four socket system that two dual sockets for the first time.

The final product to feature the latest AMD Opteron is Acer's multi-node server, which the firm claims has the best performance per watt with four hot-pluggable nodes in a 2U form.

Acer also announced it would be launching a Gateway tower server aimed at small and medium-sized businesses (SMBs) later this year when AMD's Opteron 4000 processor becomes available. The GT115 F1 claims to offer reliability and scalability whilst not being so performance driven.

As one of the big four blade server manufacturers, Fujitsu has been in this game right from the start. It was one of the first to launch a product way back in 2002.

Since then, the Primergy BX300 has been superseded by the BX600 and now we have the provocatively named BX900S1 Dynamic Cube, which focuses on high density computing, consolidation and, of course, virtualisation.

The Primergy BX900S1 takes the fight to HP's doorstep as it competes directly with its mighty BladeSystem c7000. However, the BX900S1 takes design to the next level as it offers a number of unique features that the competition will be hard pressed to match.

Standing at 10U, the BX900S1 chassis is the same height as the c7000 but scores better for server capacity as it can accommodate nine full-height or eighteen half-height blades or a mixture of both.

Redundancy is high on the agenda as there is room for up to six 2,800W hot-plug power supplies at the rear and each has a pair of removable dual-fan modules. The chassis shows off Fujitsu's 'Cool-safe' design as any bay that isn't being used is automatically covered by a self-sealing flap.

The cooling design enables air speed to be reduced, which in turn cuts noise levels. During the review we concluded that the BX900S1 was on a par with the c7000 and much quieter than IBM's BladeCenter H, which still uses two stonking great blower fans for the entire 9U chassis. The BX900S1 chassis is also divided into six cooling zones where each power supply looks after the three corresponding half-height bays at the front.

Fujitsu scores well for connection options as the new chassis has eight expansion bays at the rear - double that in the BX600 and the same number as the c7000. Fujitsu has gone further as the BX900S1 has four independent fabrics each with two expansion bays assigned to them.

This offers some interesting stacking possibilities as you can link fabrics horizontally via an internal stacking port or vertically using an external 10GbE CX-4 port on an expansion module. You can also use the latter to physically link up to four chassis together.

A good range of connection modules is already available with Gigabit and 10GbE on the table along with 4Gbps and 8Gbps Brocade fibre blades. The latter have been designed specifically for the BX900S1 but note that some of the Cisco Ethernet modules are essentially the same as those used by HP so only support up to 16 blade slots.

The processor sockets are located behind and staggered to improve cooling. Note that as the drive bays take up a fair amount of internal space, CPU0 has six DIMM sockets but CPU1 only has three. However, the BX922S1 blade will dispense with local hard disks and up the DIMM slot count to 12.

All the new blades have an embedded USB interface for booting into an embedded hypervisor and Fujitsu advised us that both ESXi and Hyper-V are supported. Network connections look good as the blades sport pairs of dual-port Gigabit adapters and support two mezzanine cards.

Fujitsu offers quad-port Gigabit, 10GbE and Emulex dual-port 8Gbps fibre mezzanine cards. A smart feature is they connect directly into the chassis fabric and unlike the competition, the extra interfaces don't have to be routed through the server blade motherboard, which reduces the length of the physical path to improve performance.

The chassis also accepts up to six of SX940 SAS/SATA storage blades. Installed in specific bays, their storage can be shared by server blades on either side or just one. Four hot-swap drive bays are provided and a unique feature is internal support for two standard PCI-e RAID cards. There's also an LTO-4 tape blade and its server connection is determined by which PCI-e slot has the HBA installed in.

For management features, Fujitsu moves up alongside HP as the BX900S1 chassis has flip-out Local Service Panel, which is very similar to HP's Insight Control Panel. It provides a handy status screen showing which chassis slots are populated and its background colour will change to denote hardware warnings or faults.

Two more slots are provided at the rear for the new management blades. These have integrated switches that connect directly to the chassis mid-plane for out-of-band access. They provide web access and the ServerView interface offers plenty of information where its home page opens with a complete overview of the chassis.

A bar chart shows current power usage across the entire chassis and power management extends to chassis power capping, which allows you to set a finite limit on maximum consumption. If this threshold is reached the chassis will apply minimum power settings across all server blades where their CPUs will be throttled back to reduce consumption.

Selecting a server blade from the ServerView interface displays a complete rundown of its hardware and provides access to event logs, server power management and basic configuration. The blade's iRMC controller can be accessed directly where you can load up its web interface and view the status of critical components and select video redirection for full remote control.

The Primergy BX900S1 in an impressive blade server with a higher server density than HP's c7000 and generally more sophisticated than IBM's elderly BladeCenter H.

Value is good and its build quality, high expansion potential and quality power management features all set a high standard for the rest to follow.

Verdict

Fujitsu’s latest blade server delivers a very high server density and plenty of innovative features. Expansion potential is extremely good, remote management and monitoring are amongst the best and there’s a lot more yet to come.

Chassis: 10U enclosure with eighteen half-height/nine full-height blade slots

Power: 3 x 2800W hot-swap supplies (max. 6)

Fans: 2 x pluggable dual fan modules per power supply

Expansion: Eight interconnect I/O module bays – two per fabric

Network: 1 x 36/12 Gigabit Ethernet switch blade

Management blades: 1 x MMB with 10/100 and 2 x Gigabit (max.2)

Management: Local Service Panel, Fujitsu ServerView Suite

Codenamed Shorty', the c3000 takes everything that makes HP's mighty c7000 great and squeezes it all into a compact 6U high chassis. Both rack mount and floor standing models are offered and in this exclusive review we take a close look at the latter.

Fortunately for us the floor standing c3000 has wheels on the chassis and the pallet it was delivered in contained a handy ramp. Be under no illusions the c3000 is built like a tank and doesn't weigh much less than one either. You can fit four full-height or eight half-height blades into the c3000's enclosure and choose from all those available for the c7000.

The c3000 runs on single-phase power and the chassis starts with four 1,200W hot-plug power supplies which can be increased to six. It uses the same cooling modules as the c7000 chassis, which are based on jet engine design principals. Take peek down the end of a fan module and you'll see that the metal fan is shaped like a turbine. They don't cost much to replace with a new module costing a shade over 100.

To one side, or in the base for the rack version, you have HP's onboard administrator module, which incorporates its Insight Display. This is a nifty pop-out operator panel and LCD screen and can be used for configuration, fault analysis and checking on general system health. A chat mode enables text based conversations to be conducted between a remote manager and local support staff.

HP offers a choice selection of server blades and these include support for the latest Series 5500 Xeons. We were supplied with the new BL490c G6, which is a half-height blade that supports a pair of 5500 processors. Eighteen DIMM sockets enables the standard 6GB of DDR3 memory to be pushed up to a total of 96GB and you can use memory mirroring as well.

The high DIMM socket count means internal space is limited for storage, but HP gets round this neatly by fitting up to two SATA SSDs over the top. You can expand capacity much further by adding an SB40c storage blade in the bay below where it connects to the server blade and adds a further six SATA or SAS SFF hard disks. RAID is on the menu too although this will require an optional Smart Array P700 controller, which slots into one the pair of mezzanine card slots.

Plenty of other mezzanine card options are available as HP offers dual-port FC, iSCSI, extra Gigabit ports or Infiniband options. Standard network connections get a boost as the new blades sport an embedded dual port 10GbE Flex-10 server adapter although to take advantage of the higher speeds you'll need to fit a virtual connect Flex-10 10GbE module at the rear.

The latest BL460 G6 half-height server blade offers a similar specification to the BL490c G6 but drops the DIMM socket count to twelve enabling it to take a couple of hot-swap SFF hard disks at the front. These are managed by an embedded P410i RAID controller and the latest blades have virtualisation as a priority as they all have an internal SD memory card slot and USB port for loading embedded hypervisors.

HP hasn't been idle since its acquisition of LeftHand Networks as the c3000 can also be configured to run its SAN/iQ software. The main concept behind SAN/iQ is clustering over Ethernet, which enables you to gather together multiple physical servers and present all their storage as a single pool. We have already reviewed this and were impressed enough to give it an IT PRO Editor's Choice award.

At the rear you have four interconnect bays available and HP offers plenty of connection options as you can pick and choose from pass-through blades, Cisco copper and fibre Gigabit switch blades and Brocade and Cisco FC SAN switches. You'll need the virtual connect modules to takes advantage of the new 10GbE offering in the latest server blades and these can be used to assign MAC addresses from a virtual pool when new blades are installed.

The level of remote management offered by the c3000 is unbeatable. The chassis, blades and all interconnect modules can be accessed directly by the Onboard Administrator browser interface. This provides wizards to help with installation and an array of graphics showing the condition of all components. It provides a system status readout showing colour coded icons for errors and faults along with details on chassis power consumption and available power.

Pick any server blade and you can go directly to the browser interface provided by its embedded iLO2 chip. This provides full remote control over all system operations including power and also offers tools for monitoring the controller or server and viewing installed components.

For more in-depth management you can use HP's Insight Control Suite, which includes the Systems Insight Manager tools. These provide enhanced browser-based remote management and monitoring and can remotely access server blades with an Insight agent installed.

The only competition to Shorty' comes from IBM's BladeCenter S, which targets precisely the same market. At 7U high it claims a greater expansion potential and higher storage capacities, but it's incredibly noisy, so if you want this in your office you'll need IBM's massive office enablement kit.

We found the c3000 much friendlier during testing as even with all six fans installed it was surprisingly quiet. We also prefer its more compact dimensions and superior build quality plus HP's management facilities really are standard setters.

Verdict

HP’s blade servers have always impressed and the c3000 is no exception. Its heavy but compact chassis offers a remarkable expansion potential with a huge range of server and storage blades to choose from, whilst the extensive portfolio of interconnect modules covers virtually every angle. HP’s management tools are the ones to beat as well and make light work of installation and deployment.

Chassis: 6U enclosure with eight half-height/four full-height blade slots Power: 4 x 900W or 1200W hot-swap (max. 6) Fans: 4 x hot-swap (max. 6) Expansion: Four interconnect I/O module bays Management: HP Onboard Administrator, HP Insight Control suite

A double-digit falloff in the fourth quarter hit shipments and revenue, with global shipments slipping 11.7 per cent and revenue diving 15.1 per cent in the fourth quarter. Strong results at the beginning of the year helped keep shipments for the year positive, at 2.6 per cent growth, but full-year revenue still fell 4.1 per cent.

The worst revenue decline was seen in Europe, the Middle East and Africa (EMEA), at 20.6 per cent, while North America marked falls of 14.6 per cent.

"The outlook for 2009 is a very challenging one, with the pattern of the fourth quarter more indicative for the level of demand in 2009 than that seen in 2008 as a whole," Kota said.

"The continued weak economic environment will cause users to be extremely cautious with levels of expenditure which will make for a particularly challenging environment for vendors," Kota added. "The server market already has high levels of vendor consolidation but the conditions expected during 2009 will increase the threat of further consolidation."

Despite the declines, blade servers saw revenue and shipment growth of 30 per cent, boosting the leaders in those systems, HP and IBM. "Blade servers were one of the few segments to achieve any growth at all in this challenging environment," said Kota. "Unix servers on the other hand fell for the fourth quarter at 10.5 per cent in shipments and 13.7 per cent in revenue for the year."

For the fourth quarter, despite diving 17.4 per cent, IBM held on to top place by revenue with 34.4 per cent of the market, followed by HP at 28.8 per cent and Dell at 10.2 per cent. By shipments, HP fell just 1.6 per cent to hold the market lead at 29 per cent, followed by Dell at 20.6 per cent and IBM with just 15.4 per cent, after its shipments fell 22.3 per.

NEC, Lenovo and Dawning were the only top ten server makers which did not see any decline in shipments.

Co-developed with Intel, the FlexPower Server is offered as a cost-effective alternative for businesses that are looking to purchase up to three servers over the coming year. It aims to offer far better value, but also adds solid upgrade and expansion paths, quality fault tolerance and some very interesting storage possibilities. This 6U chassis accepts up to six server blades, or compute modules, and has enough room for fourteen hot-swap small form factor, (SFF) hard disks, which can placed into zones and assigned to selected modules.

The compute modules has been designed specifically for this system and are all based on Intel's PAL5000 platform so are able to support dual- and quad-core Xeon processors. The price for the review system includes a pair of modules each kitted out with a 2GHz Xeon E5405 processor, teamed up with 4GB of FB-DIMM memory, which can be expanded up to 32GB. The blades also have a pair of embedded Gigabit ports and the network count can be expanded to four with an optional dual-port mezzanine card.

The storage scenario comprises two drive bays at the front each offering seven hot-swap slots for low power SAS and SATA SFF hard disks. The bays are routed through internally to the chassis midplane and link up with a storage controller blade at the rear. This has an embedded LSI SAS chip, which supports a fine range of array types including dual-drive redundant RAID-6. Selected drives are placed into storage pools from which you create multiple virtual volumes each configured with their own RAID array type.

Virtual drives are assigned to compute modules, which see them as local storage so you're effectively creating a SAN within the chassis. This offers good fault tolerance as if one compute module fails you simply reassign its virtual drive to another standby module and boot it up where it continues to provide the same services. You can also expand storage capacity as the controller has an extra SAS port for connecting external drive arrays and adding a second controller brings active/active failover into play.

The module's network ports are routed through to Ethernet switch blades at the rear and the review system came with one 10-port Gigabit module. These only support L2 switching but you can add another switch blade, which handles the second network port on each module so you can implement network failover as well. You get one management I/O module as standard, which has a dedicated network port and redundancy is supported by adding a second module.

Power is handled well as the review system included all four hot-swap supplies although with this particular module and storage configuration it is possible to run the system with just two installed. To reduce the number of moving parts the compute modules don't have any onboard fans and all cooling is handled by two large hot-plug fan modules at the rear and a third, smaller one underneath the drive bays.

With SMBs as the target market NEC has focused closely on ease of deployment and management and first contact with the Modular Server Control (MSC) web interface indicates this has largely been achieved. The console home page provides a dashboard showing power status, enclosure, drive and CPU temperatures, system health and detected problems. Using the row of tabs at the top you can bring up graphics of the front and rear of the chassis showing all installed components and a table below provides more detail about each one

Full remote access is provided for the compute modules where you can power them on and off and use the KVM over IP feature to gain full remote control. The latter makes this system even better value as some other vendors only offer remote control as an option with their management solutions. The switch blades also have their own web interface, which can be accessed from the console enabling you to monitor network activity, configure ports and create VLANs.

We found storage easy enough to configure from the MSC and created storage pools with the available drives and assigned virtual drives to our two resident compute modules. We had no problems installing an OS on them as we used the local optical drive on the system running the MSC as a virtual boot device. To move a virtual drive to another module you call up the Transport option where you can take the drive offline and assign it to another module. It's easy enough to keep track of all storage as the MSC provides a slick flow-chart style graphic that shows clearly what each storage pool has in it and which compute modules the virtual drives are assigned to.

One criticism we have of MSC is its power monitoring facilities as we found these to be highly inaccurate. With all four power supplies connected and both compute modules powered down our inline power meter showed a draw of 300W whereas the MSC reckoned this was only 90W. With both modules running, the meter reported an increase to 506W but the MSC only showed a total draw of 270W.

Administrative access to the entire chassis can be strictly controlled with user lists and for each one you can decide precisely, which components and features they can access. Plenty of notification options are provided and these are controlled with policies that determine the logging levels and who should be alerted based on the severity of an event.

SMBs that thought blade server technology was beyond their means should check out NEC's FlexPower Server as it looks a very interesting alternative to the mainstream products. In the current financial climate many businesses large and small will undoubtedly be looking closely at their IT budgets and this system offers a cost-effective solution for consolidating systems and reducing overall expenditure.

Verdict

The FlexPower Server brings blade servers to the masses as it offers a well built, high density processing package at a reasonable price. The virtual storage mapping feature is quite unique and it offers a lot of flexibility. Fault tolerance is also particularly good, there’s plenty of room to expand and the remote management tools make light work of deployment.

Chassis: 6U rack enclosure with six compute module slots

Power: 4 x 1050W hot-swap supplies

Fans: 3 x hot-swap modules

Storage Controller Blade: 1 (max. 2) supports RAID0, 1, 1E, 10, 5, 6

Storage: 8 x 147GB Fujitsu 10K SAS SFF hard disks in hot-swap carriers (max. 14)

Switch blade: 1 x 10-port L2 Gigabit Ethernet with web management (max. 2)

Management: 1 x Management blade with 10/100 Ethernet, serial port (max. 2)

Compute Modules: 2 (max. 6)

CPU: 2GHz Xeon E5405

Memory: 4GB 667MHz FB-DIMM (max. 32GB)

Storage: Embedded LSI SAS1064E

Network: 2 x Gigabit Ethernet (max. 4)

In IT PRO we also took an exclusive look at IBM's BladeCenter S and liked what we saw here as well. The latest Modular Server from transtec certainly looks capable of competing on price with the blue chips and in this exclusive review we see how well it stacks up against them.

Intel's influence behind the scenes with this system is strong. It introduced its Modular Server for three markets: for SMBs planning on consolidating IT services onto a single system, for those looking for a single platform with plenty of expansion potential, and for businesses with virtualisation as a priority. The Modular Server is positioned as an entry level blade solution with ease of deployment and management high on the agenda.

This 6U rack mount chassis has six horizontal slots at the front for the compute modules. There's a good choice on offer as the customised Intel motherboard is based on the PAL5000 platform and so supports dual- and quad-core Xeon DP processors. The price for the review system includes a pair of modules and each was kitted out with 8GB of memory, which can be expanded to 32GB. The pair of embedded Gigabit Ethernet ports can be doubled with an additional dual-port mezzanine card.

For storage, the Modular Server takes a similar tack to IBM's BladeCenter S, which uses two storage modules each holding six 3.5in hard disks that can be split up into zones and assigned to selected server blades. However, the Modular Server looks more versatile as the two bays at the front each support up to seven SFF hard disks.

Both bays are linked to the chassis midplane, where they are routed through to a storage controller blade at the rear, which supports a good range of arrays including RAID-6.

I/O blade options are currently restricted to Gigabit Ethernet and the base system comes with one switch blade and a spare slot for a second. These manageable L2 switches each have ten Gigabit uplink ports and twelve internal ports, which are linked to the compute modules via the chassis midplane. The chassis comes as standard with one management module, which provides remote access to the chassis and all components and you can add a second for redundancy.

For remote management, the Modular Server Control (MSC) web interface provides easy access to all components. A dashboard view reveals power status and consumption, temperatures for the enclosure, drives and CPUs, system health and alerts to problems. A row of tabs above enables you to swap to a graphic of the front and rear panels showing installed components and selecting one provides more detail about them in a table below.

The accuracy of the power monitoring graphs is questionable as the chassis with no modules powered up was shown by the MSC as pulling 102W whereas our inline power meter registered 225W. With one blade fired up this rose to 240W in the MSC whereas our meter showed 370W. With both blades running MSC registered 384W but our meter showed 517W.

The storage arrangement is very flexible as you select physical drives and place them in storage pools. Within these you create virtual drives each with their own RAID array type. Only those arrays that can be supported by the number of selected drives are made available for selection and the array is assigned to virtual drives and not the storage pool. This enables a storage pool to support multiple arrays and each virtual drive can be a different RAID array as well. Your next job is to assign virtual drives to selected compute modules where they will see them as normal local storage.

Essentially, this enables an internal SAN to be created. A stand-out feature is that if a module fails you can re-assign its virtual drive to another one and boot it up with minimal disruption. The Transport option is provided for this function where you take selected drives offline and move them across. External expansion is possible as the storage controller blade has an external SAS port and adding a second storage blade creates an active/active failover scenario.

Each blade can be controlled locally by attaching a monitor and USB input devices to them but the MSC provides KVM over IP access so you can control power to switch them on and off and remotely view and control their BIOS and OS. It's easy enough to see your storage layout as the MSC Report tab provides a smart flowchart diagram showing drive bays, installed drives, storage pools, virtual volumes and which modules they are assigned to.

Access controls for chassis management can be implemented by creating new users and deciding precisely which components they can interact with. Firmware upgrades can easily be applied to the entire chassis and all installed components. Usefully, once the process is completed it will only automatically reboot components that will not cause any downtime and leave the rest to be manually rebooted when convenient.

Stack up transtec's Modular Server against the SMB offerings from HP and IBM and it's clear there are distinct advantages. It's not as well built as HP's c3000 but it's a lot quieter than IBM's BladeCenter S, which will need the 11U high enablement kit for deployment in an office. HP's and IBM's remote management features are more comprehensive but the Modular Server's management tools are nothing to sniff at, it looks comparatively good value and its storage facilities make it very versatile.

Verdict

SMBs looking to use blade servers to consolidate IT services on a single platform have had a limited choice, with only HP and IBM on their list. transtec’s Modular Server offers a very interesting alternative as it brings the entry point down to a more affordable level and offers a superior storage proposition to the establishment.

Chassis: 6U rack enclosure with six compute module slots Power: 2 x 1050W hot-swap supplies (max. 4) Fans: 2 x hot-swap Storage Controller Blade: 1 (max. 2) supports RAID0, 1, 1E, 10, 5, 6 Storage: 4 x 73GB Seagate Savvio 15K.1 SFF SAS hard disks in hot-swap carriers (max. 14) Switch blade: 1 x 10-port L2 Gigabit Ethernet with web management (max. 2) Management: 1 x Management blade with 10/100 Ethernet, serial port (max. 2) Compute Modules: 2 (max. 6) CPU: 2 x 1.6GHz Xeon E5310 Memory: 8GB 667MHz FB-DIMM (max. 32GB) Storage: Embedded LSI SAS1064E Network: 2 x Gigabit Ethernet (max. 4)

With the BladeCenter S, IBM is primarily targeting the high growth small to medium business market and particularly those looking to consolidate and centralize their IT services. It also has its eye on large enterprises with geographically distributed locations such as branch offices and users where blade server technology has traditionally been too expensive or impractical due to size of the enclosures and operational noise levels.

Standing at 7U high the BladeCenter S is slightly taller than HP's c3000, which is a more compact 6U. Features that make this unique in the BladeCenter family are the dedicated areas for two storage blades and up to six server blades. The c3000 initially looks better on paper as it accepts four full-height or eight half-height blades or a mixture of both but the BladeCenter's storage facilities look a lot more versatile.

In the S chassis you have two dedicated blade slots at each end that accept IBM's storage modules. These each hold up to six 3.5in SAS or SATA hot-swap hard disks and connect directly into the chassis' midplane where they are placed in zones, enabling you to dish out storage to selected server blades. There are plenty of zone options and the review system came with a default configuration, which handed three drives in each storage module to the server blades in slots one and four. Other predefined zones are available or you can create custom ones and the server's embedded LSI Logic RAID controllers see their allotted storage as local.

Although storage blades for the smaller SFF hard disks aren't currently available IBM advised us it expects to support them later this year. Note also that the modules are not offering a shared storage solution as each drive can be mapped only to one server. The storage blades don't limit expansion potential as IBM's SAS Connectivity Modules can increase this massively. These slot into the rear expansion bays and offer four SAS ports to which you can direct attach IBM's external SAS disk arrays and external tape drives.

IBM definitely wins for storage options as its SAS I/O modules make the BladeCenter S quite unique. HP's SB40c storage blade supports six hot-swap SAS or SATA SFF drives and has it own integral Smart Array RAID controller but it only connects to the server blade it is installed next to, effectively becoming a simple DAS solution.

/*-->*/

/*-->*/

The high operational noise levels of enterprise blade servers makes them totally unsuited to the office but for the c3000 we found general noise levels with a half-populated chassis to be surprisingly low mainly due to the fact that it uses the same turbine style fans as the c-Class. We found that the BladeCenter S was far too noisy in the lab even with only two server blades up and running but IBM offers its office enablement kit to remedy this.

It comprises a rack chassis mounted on wheels that has an acoustic baffle at the back. The cabinet is 11U high enabling it to accommodate other systems up to 4U high and keep them quiet as well. If you want peace and harmony and a BladeCenter S in your office all at the same time then this kit is a must.

The BladeCenter S is designed to be easy to install and we can heartily agree. Network the AMM (advanced management module), point a browser at its default address and a wizard takes you through key functions such as setting up and securing remote management access, entering IP addresses for the switch blades and storage modules, adding email notification addresses and configuring storage zones.

The home page provides a complete overview of the chassis and installed blades and although it's not as well designed as HP's Onboard Administrator interface it does provide easy access to the chassis and installed blades. The system status page keeps you posted on the condition of all server blades, I/O interconnect modules and drives in the expansion modules and power settings for each blade can be accessed directly.

For power management AMM keeps you posted on consumption, the total available and what's left in the tank. An acoustic mode attempts to throttle a blade's processors back to keep the cooling fans within accepted noise levels. IBM's Director is its answer to HP's Systems Insight Manager and provides tools for managing IBM servers, workstations, laptops and blade servers.

Along with a wide range of features including enhanced inventory, system monitoring and software distribution Director also offers power management. Renamed from Power Executive to Active Entergy Manager, this component offers real time power monitoring of entire BladeCenter chassis and individual blades and provides power capping, which enables CPU clock speeds to be modified to reduce power consumption.

Businesses looking to implement blade server technology in the office have two straight choices - HP's BladeSystem c3000 or IBM's BladeCenter S. The c3000 is more compact and quieter and has superior server and power management facilities. The BladeCenter S offers an equally extensive range of blades and I/O interconnect options plus superior storage options but if it's going into an office then the enablement kit will be a mandatory requirement.

Verdict

A very well specified blade server that'll give HP's BladeSystem c3000 a good run for its money. Remote management and monitoring capabilities come second to HP but for storage features and expansion the BladeCenter S offers some unique options.

Chassis: 7U rack enclosure with six server blade slots and two storage module slots Power: Four hot-swap supplies Fans: Four hot-swap modules Expansion: Four I/O module bays Management: One advanced management module Other: Office enablement kit, serial pass-thru module, media tray with DVD-RW, 2 x USB Blade 1: 2 x Intel Xeon E5320 1.86GHz, 2GB 667MHz FB-DIMM memory Blade 4: 2 x Intel Xeon E5320 1.86GHz, 2GB 667MHz FB-DIMM memory Storage Expansion Modules: Each with 6 x 73GB IBM 3.5in SAS hard disks in hot-swap carriers

Quoting research from AMI Partners, HP says that more than 60 per cent of midsize companies will be looking to deploy technology to help with customer relationship management and secure remote access.

To address this market, HP has released the ProLiant BL260c G5 blade server and the DL120 G5, an entry-level rack optimised density line server.

Both offer enhanced CRM configuration options, Linux-based management tools and remote access technology.

These come in the form of Microsoft Dynamix CRM software or Oracle Siebel CRM Professional Edition, HP's Insight Control Environment management tool, and the HP Secure Remote Access software.

HP claims that the BL260c is the most energy efficient blade server in the world, being 20 per cent cheaper and 64 per cent more power efficient than any other blades on the market.

"Through extensive research and customer interaction, HP has gained valuable insight into the needs of midsize business customers," said Urs Renggli, director of worldwide small and midmarket business for HP's Technology Solutions Group. "As a result, we can deliver customised, affordable and effective technology designed specifically for them."

The servers are designed for use with the HP Bladesystem c3000 and c7000 enclosures and are suitable for branch offices and compact remote sites with virtual machine environments.

It's offered with either Intel Xeon 5200 and 5400 series processors, supports up to 24GB of memory, has redundant network connections to maximise uptime and features RAID support.

The DL120 G5 is based on a single dual-core or quad-core Xeon processor, and is backed with 800MHz PC2-6400 memory. It has two PCI Express slots and can be upgraded with HP Serial Attached SCSI (SAS) Host Bus Adapters (HBAs) and Smart Array Controllers and has optional LO100c remote management capability.

Both servers are available immediately, with pricing for the HP ProLiant BL260c G5 starting at 665 and the HP ProLiant DL120 G5 starting at 509.

Supermicro is the power behind the throne as the HXBL200 is its new SuperBlade system. This Tiawanese system manufacturer has been threatening to deliver a blade server for a number of years but has always held back. The reason it has launched now is it wanted to offer a system that it truly believed was a cost-effective alternative to standard rack servers. With Viglen's Xeon starter bundle costing a shade over 6,000 it certainly looks to have achieved this as this particular package includes not only the chassis, a Gigabit switch blade and a pair of power supplies but also a quad-core E5365 Xeon server blade as well. The AMD startup package looks even better value as this includes one server blade with no less than four quad-core AMD 8347 processors. Add in the fact that the 7U chassis has room for ten blades and this equates to no less than 960 Opteron cores in an industry standard 42U rack - a density no other blade server manufacturer can currently match let alone beat.

General build quality is very good although the cooling arrangements leave something to be desired as the power supplies are also tasked with these duties and are equipped with large internal radial fans. At the rear you have room for up to four supplies with the price for the startup packages including a pair of 2000W modules. Adding two more gives you good power redundancy and Supermicro offers a choice of 1400W, 2000W and 2500W options. It claims these are all highly efficient and has shown some forethought as the high output model is designed to supply enough juice for the next generation of Xeon MP server blades. The supplies are clearly capable of shunting large volumes of air through the chassis but we reckon the turbine style cooling modules on HP's BladeSystem c-Class are a far more elegant, and quieter, solution.

The chassis has four expansion slots at the rear that accept Supermicro's pass-through, Layer 2 Gigabit switch and InfiniBand blades. The Gigabit switches provide ten internal and ten external network ports and with two installed you can use both network ports on each server blade. What you don't get at the moment are options for fibre channel links such as those offered by HP's Brocade blades. For the Infiniband option you need to install an additional mezzanine card on each server blade and Supermicro also plans to offer a 10-Gigabit option that uses the same arrangement. A planned future option that will make the HXBL200 more unusual is the possibility of adding expansion modules that accept PCI-e cards which link up with the two 8X PCI-e buses on each server blade.

For server blades Viglen currently offers Xeon dual- and quad-core processors or AMD's quad-core Opterons. The Xeon blades accept a pair of 3.5in. hot-swap SATA drives and the embedded RAID controller supports stripes and mirrors. Alternatively, you can plump for the Xeon blade that supports up to six 2.5in. SFF SATA drives and adds RAID-5 arrays to the mix. What you gain in processing power for the AMD quad-socket blades you lose in the storage department as the four sockets on the blade motherboard only leave enough room for a couple of fixed internal 2.5in. SFF SATA drives to be fitted.

Remote management is on the cards as the chassis has two more expansion slots at the rear that accept Supermicro's chassis management modules (CMMs) which provide web browser access to the system. The interface is well designed and provides good levels of access to the chassis, server blades and all interconnect modules. Local devices on the management system can be used as virtual boot devices for selected blades and we liked the fact that remote control comes as standard - HP charges extra as this is an optional feature on its Systems Insight Manager software. During testing we found it easy enough to deploy and manage the HXBL200 as the blades can also be accessed locally using the supplied cable as well as remotely over KVM.

Whereas HP and IBM both offer OS and application deployment tools with their respective management software Viglen has no need to do this as it will supply the system preconfigured with your choice of OS. The system can also be accessed over IPMI and Supermicro has revamped its IPMI View 2.0 software which now delivers a very tidy interface to accommodate the HXBL200. Graphics are provided to show the status of all installed blades plus power supplies, CMMs and expansion modules. You can also keep track of the fan speeds in each power supply and chassis environmental values as well. For general blade server management the bundled tools do the job nicely but you'll need to look towards HP's or IBM's blade servers if you want sophisticated chassis and blade power controls.

This blade server has been a long time coming but Supermicro has made the wait worthwhile. Viglen's HXBL200 may not offer the same level of expansion options or remote management capabilities as HP's BladeSystem c-Class but it certainly wins out for sheer value and a quite remarkable processing density.

Verdict

Viglen delivers a well built blade server with value high on the agenda. It's not as sophisticated as equivalent systems from HP and IBM but for your money you're getting a lot of hardware and the highest processing density currently on the market making this a very cost-effective alternative to standard rack servers.

As reviewed:

7U enclosure

10 blade slots

2 x 2000W hot-swap power supplies (max 4)

CMM blade with KVM

10/100 Ethernet

Monitor and USB ports (max 2)

10-port Gigabit Ethernet L2 switch blade

Xeon blade - 2 x 3GHz Xeon E5365

16GB 667MHz FB-DIMMs expandable to 32GB, 500GB Western Digital RE SATA hard disks in hot-swap carrier, 2 x Gigabit Ethernet.

AMD startup package option:

Same enclosure configuration with 1 x AMD Opteron blade - 4 x 1.9GHz Opteron 8347

16GB 667MHz DDR2 expandable to 64GB

80GB Seagate Momentus SFF SATA hard disk

2 x Gigabit Ethernet

£6,719 exc VAT. 3yrs on-site warranty

Not a lot of people know this but NEC Computers has had a presence in the blade server market for over four years. However, it has never promoted its products aggressively - certainly not the to same level as IBM which is currently spending a king's ransom advertising its BladeCenter H in any form of media that comes to mind. NEC's first attempt came with its Express5800/420Ma and here we take an exclusive look at its successor - the provocatively named SIGMABLADE.

Standing at 6U high, the SIGMABLADE-M chassis is 2U higher than its predecessor and supports up eight server blades. The first thing of note is its rude construction as although it's undeniably a sturdy beast it lacks any of the elegance of HP's BladeSystem c-Class (see https://www.itpro.com/servers/labs/95397/hp-blade-system-cclass.html). NEC also offers a SIGMABLADE-H chassis which stands at 10U and along with the 120 range can also use NEC's 140Ba-10 double height blade which supports Intel's 7300 Xeon MP quad-core processors. Underneath the server blade slots are small bays for up to four hot-plug power supplies. These are an unusual design as at nearly 80cms long they run the entire depth of the chassis. A raised section half way down mates with the chassis' power bus and the far end presents a standard three pin socket.

At the centre of this lower section is a panel combining a DVD drive, a couple of USB ports and an LCD display. It's not as clever as HP's Insight Display as it only provides basic chassis and blade information and allows you to swap the KVM port and integral DVD drive assignment to a selected server blade. However, it can be flipped out to reveal a couple of slots behind that take NEC's enclosure manager (EM) cards. These offer web access to the chassis and adding two EM cards provides redundancy. They can each be accessed directly via their serial or Ethernet ports or you can use the rear mounted module which also provides KVM access to the server blades.

At the rear you have room for up to five fan modules and connectivity options are reasonably good as below the fan bays are six slots for which NEC offers a choice of hot-plug modules. The review system was supplied with a basic Gigabit pass-through module which links up with both Ethernet ports on each blade. You also have a 16-port Layer 2 Gigabit switch, an 8-port 4Gbps FC module and an FC pass-through module as well. Server blade choices are more modest than those from HP and IBM as NEC currently offers three choices of Xeon based blades supporting dual- and quad-core processors. The 120Bb-6 blades in the review system each had a pair of 1.6GHz Xeon 5110 processors teamed up with 1GB of memory. Storage is handled by a RAID mezzanine card at the rear which is cabled through to a small backplane servicing the pair of hot-swap SFF SAS drive bays. It's a tidy enough solution but the drive bay does sit on top of the DIMM sockets and will need to be unscrewed and removed if you want to upgrade memory. NEC also offers a diskless blade which is designed to boot from a SAN and is aimed at storage consolidation applications. Of more interest is its storage blade as from a hardware perspective this is similar to HP's AiO SB600c solution. The 120Bb-m6 combines a standard diskless Xeon blade with a second that supports up to six hot-swap SFF SAS hard disks.

For remote management the EM card web interface provides a basic status screen showing the server blades, fans and power supplies and you can directly access each server blade's own web interface where you can perform power controls, reset them, shutdown their OS, remotely control them and assign virtual boot media.

Next up is NEC's ESMPRO utility which provides remote access to the server blades and starts off with a discovery routine that scans the network and lists all SNMP devices. Once you've installed a local ESMPRO agent then any NEC server can be queried directly from the map for hardware inventory and performance graphs of selected components. Each server blade has an embedded IPMI chip which can be accessed using the bundled DianaScope utility. This offers a basic web interface which allows you to monitor server activity, remotely control them using the VNC software, shutdown or restart their OS and schedule power cycles.

Stacking up NEC's remote management tools against the likes of HP's Onboard Administrator web interface and Insight Control Center Edition shows clearly what you're missing as the HP alternative is much slicker. Furthermore, NEC doesn't provide the same power management and monitoring tools as HP and IBM. Although not supplied with the review sample, NEC does also offer its SigmaSystemCenter administration software which is designed to streamline OS and patch deployment across multiple blades, automate server recovery operations and balance workloads across selected servers.

Having seen nearly every blade server on the market we can safely say that HP's BladeSystem c-Class is the benchmark by which we judge all others. However, this quality does come at a price and this is one area where NEC's SIGMABLADE offers a strong alternative as the system on review looks particularly good value. The new storage server blade is a welcome addition and component redundancy is extensive but we did find the standard administration facilities unsophisticated and lacking in power management tools.

Verdict

NEC's new blade server scores very well on value and expansion capabilities and supports a reasonable range of Xeon server blades. Redundancy options are plentiful but build quality is unsophisticated and management and monitoring features aren't a patch of those offered by HP's BladeSystem c-Class.

6U rack chassis - 8 blade slots, 2 x 1250W hot-plug power supplies (max. 4), 4 x hot-plug fans (max. 5), EM card with Ethernet and KVM (max. 2), six module expansion slots, 16-port Gigabit pass-through blade. 2 x 120Bb-6 server blades each with: 2 x 1.6GHz Xeon 5110, 1GB 667MHz DDR2 upgradeable to 16GB, LSI Logic MegaRAID card, 2 x 73GB Seagate Savio 10K.2 SAS hard disks in hot-swap carriers. NEC ExpressBuilder software suite bundled. Warranty - 3 years on-site.

In a bid to both extend the market for blade systems and take 10 Gigabit Ethernet blade systems mainstream, the companies are anticipating strong demand for 10 Gigabit Ethernet blade servers for virtualisation and High Performance Computing, but also bandwidth-intensive apps such as IPTV and massively multiplayer online gaming.

The BLADE 10Gbit/sec switch provides 20 10 Gigabit Ethernet ports, with 14 internal to the chassis and six external, for connection to the network. The BLADE switch is used with IBM BladeCenter via a dual-port 10 Gigabit Ethernet expansion card from NetXen, which can support multi-protocol offload.

"We are taking datacenter performance and affordability to the next level for enterprise customers and paving the way for mainstream adoption of 10GE blade systems," said the President of NetXen, Dave Pulling.

The BLADE Nortel 10Gbit/sec Ethernet Switch Module is priced at $9,799 and the NetXen 10 Gigabit Ethernet Expansion Card is $899, with immediate availability from IBM.

The announcement, from the Blade.org summit, coincides with the first anniversary of the organisation. Set up in 2006 by the likes of Intel, IBM, Citrix and Nortel, it has been trumpeting the investment made in the blade market - $1bn (560m) via more than 50 venture capital companies.

Intel and IBM first went public with design specs for the BladeCenter platform back in September 2004.

They decided that companies that base their designs on the BladeCenter platform would not have to pay any royalties or license any patents. The only charges the partners make is if manufacturers need the assistance of their engineers.

According to IDC it is the industry's fastest growing server market segment predicted to reach $11bn (6.6bn) in revenue by 2010.

A Blade Server consists of a motherboard with single or multiple processors, memory, a network connection and sometimes storage. The ultra-thin devices are intended for use by large computing centres where space is at a premium.