Most virtualization goofs center around lack of planning and forethought, they said.

For example, on the storage front, perhaps the biggest mistake is to start a virtualization project without already having a SAN or shared storage in place, said Robert Passmore, a Gartner research vice president. “They get 75% of the way through a virtualization project and they realize they need a SAN, but by that point they don’t have the budget,” he said. “It’s a real disaster.”

Similarly, most virtualization newbies don’t anticipate the increase in demand that virtualization will place on their systems, nor the speed at which growth will occur.

“Virtualization introduces speed, and most processes aren’t ready for speed,” said Tom Bittman, another Gartner research VP. And because virtualization tends to remove artificial obstacles to provisioning new workloads, demand for services tends to double in highly virtualized environments, he said. To avoid being overwhelmed, newcomers to virtualization should consider implementing lifecycle management, as well as chargeback or showback.

Other rookie mistakes involve failing to treat the virtualization layer with enough gravitas.

In their enthusiasm for VMware, many IT managers make the mistake of moving too quickly to the latest and greatest release of the platform, said Passmore, oftentimes before ecosystem such as backup and management software are available.

And security of the virtualization layer is often overlooked, added Neil MacDonald, vice president and distinguished analyst. “These issues are overlooked because people say ‘nothing’s different,’ when really, a lot is different,” he said.

To avoid security problems, MacDonald suggested IT managers elevate virtualization to the same layer as the operating system. “Treat it like an OS,” he said, with all the attending hardening, patch management and compliance processes.

Finally on the desktop front, an overarching mistake is to fail to fundamentally rethink desktop support, said Mark Margevicius, vice president and research director. When it comes to the desktop, most organizations are built around a distributed device, he said, but “everything changes when you virtualize the desktop,” for instance, support processes, refresh rates, and budget allocation, to name a few. Without thinking desktop virtualization through up front, “you almost have more problems than the problems you solved with it.”

Instead of showing off the new system – which they refer to as “the new movement” – with some demonstrations, we watched 90 minutes of Cisco’s CEO John Chambers and partners Intel, BMC, Microsoft, EMC and VMware congratulating each other on being masters of the universe. Good thing I had that barf bag nearby.

After Cisco and its partners were done talking about how revolutionary this new system is and how much they love each other, one reporter basically asked, where’s the beef? “We have been hearing about the California server for weeks now, but you haven’t mentioned anything about a server. Is this announcement related to that?,” he asked.

Before Chambers let his trusty engineer answer the question, he thanked all of his partners again. The Cisco engineer then reiterated  their strategy with this system while carefully avoiding the term “blade server” because the system is more than just that. And round, and round we went.

Bottom line, the system is a chassis full of Cisco UCS B-Series blades bundled with networking, storage and virtualization features. Take the pieces apart, and you have Cisco’s first blade servers. Some people may also have found it interesting that the Intel Nehalem-based servers come in both full and half depth options, so you can pack a ton of the half-depth boxes into a chassis (assuming they don’t throw off crazy amounts of heat).

So the fact that Cisco’s talking-head-style news conference was absolute torture doesn’t make the system itself any less interesting from a server market perspective. We already know their networking stuff works, so they really just have to prove themselves with some solid server engineering to compete with the existing x86 providers. (Cisco, I know you say you aren’t competing with those guys, but you are).

And in many ways, Cisco has come full-circle by introducing a server, said Anne Skamarock, a research director with the analyst / consulting firm Focus.

“When I worked at Sun Microsystems back in the mid-1980′s they debated becoming a Cisco putting intelligent switches (read: specialized servers) in the network. So in a very real sense, Cisco has been building servers for years – servers designed specifically for the work of switching,” Skamarock said. “If you think about it, the first “blade servers” were produced in the networking space years ago adding a form factor for multiple switches from the horizontal to the vertical.”

Cisco also talked about how much this system will save companies because it “radically reduces the number of devices and the required setup, management, power/cooling, and cabling,” but they didn’t talk about the acquisition cost. A Cisco spokesperson said they can not release any pricing details until April, but I am betting it is not a small number.

Even so, if Cisco has engineered a solid server and the system as a whole proves to be of good value, the Unified Computing System concept will catch on, but we aren’t sure when these systems will actually hit the commercial market.

And I’m sure server vendors like HP, Dell and IBM will follow suit with their own “me-too” unified systems similar to Cisco’s. Actually, those companies may even end up using their top networking partner for the plumbing. After all, in terms of virtualization, Cisco has come up with important technologies like VLANs and VSANs, which are now industry standards.

The way I see it, by creating this “new” market of Unified Computing Systems, Cisco is setting itself up for success in both the networking market and the server market.

Yup, all the operations for the largest sporting event in America ran on IBM’s smallest systems, the BladeCenter S, which is similar in size to a briefcase.

In addition to the big game, next month, IBM will officially announce that every one of the NFL’s 32 teams will be standardizing on BladeCenter S, according to Alex Yost, VP of IBM BladeCenters.

The BladeCenter S is actually designed for small to medium sized businesses that don’t have their own data center and need a compact, all-in-one piece of equipment, according to IBM. It is esentially a data center in a box that contains up to six IBM BladeCenter servers, 9 terabytes of local shared storage and networking components. Everything in the BladeCenter chassis is redundant – power, switching, cooling, and storage, so there is no worry about failures, either, Yost said.

As it turns out, this little data center box on wheels has made life for the NFL’s IT team a lot easier; in the past, for every Super Bowl, the NFL’s IT staff have had to lugg all the necessary servers, storage and networking to the event site and set up an entire data center within just a couple of weeks, said Jonathan Kelley director of infrastructure computing for the NFL.

As a long time IBM BladeCenter H customer that trusts IBM equipment, the NFL contacted IBM last year for some help setting up for the 2008 SuperBowl in Arizona - which this New England fan dares not discuss – and heard about the BladeCenter S.

“When the NFL came to IBM to help them set up multiple data centers for last years Super Bowl, our IBM BladeCenter S was still about three weeks away from deployment, but the NFL was confident enough in IBM to use a brand new type of server, and it went off without a hitch,” Yost said.

To support the operations of Super Bowl 44 last night, the NFL used four BladeCenter S chassis with eight, quad-core processor powered blade servers in them and about eight virtual machines running on each server to support security and credentialing for 60,000 temporary employees and around 11,000 media personell.

They also deployed about 300 PCs, wireless networks, and other necessary computing functions using IBM’s BladeCenter blades, said Joe Manto VP of Information Technology for the NFL.

“The operations at the NFL are all supported IBM blades. We chose them because their technology has proven itself,” Manto said. “These servers are almost over-engineered for what we do with them, and they are reliable.”

(The NFL wouldn’t disclose which CPU vendor they use, or name a specific virtualization vendor; they said they use a mix of virtualization vendors, but IBM reported the NFL uses VMware Inc.)

The BladeCenter S enclosure also has extra space for UPSs or other components that might need to be added, and plugs into a regular wall outlet and Ethernet connection.

“It is great for events because it is portable and can be configured at a partner site then shipped to the right location,” Yost said. “Also, their own storage can be connected to the Bladecenter S. It is super quiet and could be placed in office environments without worrying about the noise, and both the front and the back doors of the chassis lock.”

A tyical deployment is in the $15,000 range, the chassis itself is a few thousands dollars. Cost depends on the number of servers in it and other configurations, Yost said.

Urquhart posits that the combination of big news signals entry into a new era for data centers:

The long and the short of it is that we have entered into a new era, in which data centers will no longer simply be collections of servers, but will actually be computing units in and of themselves–often made up of similar computing units (e.g. containers) in a sort of fractal arrangement. Virtualization is key to make this happen (though server virtualization itself is not technically absolutely necessary). So are powerful management tools, policy and workflow automation, data and compute load portability, and utility-type monitoring and metering systems.

Recently, I discussed the concept of a “hybrid” model of cloud computing with Steve Brodie, CMO of SkyTap, and Ian Knox, Director of Product Management at SkyTap. The company announced the launch of their API that allows the transfer of existing dynamic environments to the cloud. The main focus of the SkyTap API is to enhance software quality testing, and our colleagues at SearchSoftwareQuality.com discussed the implications of the announcement for that market (Virtual environments ease software development, testing) last week.

This week, SkyTap announced that they were joining the vCloud initiative, bringing their hybrid model into the mix. The company offers a different service than other cloud hosting providers, in that their API allows users to spin up their existing infrastructure into the cloud, rather than having to build applications within the cloud.

The company explained the advantages of this model in their VMWorld press release:

Using a ‘hybrid’ cloud computing model, organizations now have a way to rapidly realize the benefits of ‘cloud economics’. The hybrid approach provides a low-risk adoption path to cloud computing and can deliver outstanding ROI compared to dynamic environments that fluctuate dramatically and are expensive to administer. In a hybrid model, companies may run their production applications onsite while conducting all their development and testing in the cloud. This enables on-demand scaling of test environments as needed and eliminates the cost of underutilized hardware. This approach also allows organizations to benefit from the management and automation capabilities of a fully automated hosted virtual lab solution, leading to huge productivity increases.

Knox ran me through a quick demonstration of how the company’s Virtual Lab works, and I was pleasantly surprised with the relative ease with which a user could connect in a virtual classroom or testing environment. The virtual lab is essentially a pool or library of hosted virtualized infrastructure that allows organizations to scale up and down lab resources as needed. Sometimes I find that the cloud is confusing to the less spatially-oriented among us, but the company’s website has a great illustrative graphic that shows how it works:

The folks at SkyTap are quite optimistic, dare I say certain, that cloud computing is the future of IT.

“It’s kind of inevitable,” said Knox. “It’s going to happen. The huge capacity headache no longer has to be borne by every company out there. With companies experiencing high pain right now, solutions like this make it so easy to get going.”

Users who have taken advantage of this easy way out of the pain concur.

“The promise of cloud computing is enormous, but with most cloud services providers you need to buy into their way of doing things from the start,” said Peter Horadan, of Admit One Security. “Skytap, on the other hand, does virtualization the same way most IT teams are used to doing it. Teams can keep their same processes and skills and use Skytap Virtual Lab as an extension of their existing environment as needed.”

If you’re interested in what is going on at VMWorld, we have it covered.

Unisys’ new ES7000 Model 7600R Enterprise Server is based on the new six-core Intel Xeon processor 7400 series. It has 16 sockets providing up to 96 processor cores. According to Unisys, the 7600R is designed for database and online transaction processing environments, large-scale consolidation and virtualization initiatives and business intelligence deployments with Microsoft SQL Server.

Model 7600R can support consolidation of 64 SQL Server databases into a single four-socket, six-core Xeon processor configuration – with 24 total processor cores – which Unisys claims is better than a commodity server farm of 64 dual-socket, single-core Xeon processor servers with 128 total processor cores, while using less disk and providing better response times.

The new server also supports VMware ESX Server and Microsoft Hyper-V, and supports dynamic partitioning so users can add more processor, memory and I/O resources on the fly without disrupting system operations. Unisys plans to introduce secure partitioning in the first half of 2009, which provides partitioning capabilities at the processor core level.

Prices for the ES7000 Model 7600R range from $26,430 to $135,000. Unisys will exhibit the ES7000 Model 7600R at VMworld 2008 in Las Vegas, Sept. 15-18.

Unisys business services
Unisys also announced new Business Assurance Services that help companies evaluate the cost and benefits of disaster recovery products, reduce the time it takes to deploy the best ones and reduce operational costs by improving resource utilization.

“We are vendor-agnostic and will implement whichever technology is best for the client. It could be a Unisys product, or it could be from another vendor,” said Jody Little, vice president of solutions and services at Unisys.

The Unisys Business Assurance Services, using discovery processes and tools developed with support from Unisys partner GlassHouse Technologies, include the following:

• Unisys Disaster Recovery Architecture Service, which provides a methodology to build application and data disaster recovery capabilities.
• Unisys Backup Modernization Service, which helps clients select new technologies and services to support backup environments at both core and remote sites.
• Unisys Data Protection for Backup Service, which helps clients improve backup and restore operations for business information, reducing costs by improving utilization of assets. Unisys experts also make vendor-independent recommendations and create a prioritized action plan

Unisys has also added new management software components to its Infrastructure Management Suite, which automates and orchestrates management of a real-time IT infrastructure. More information can be found on the Unisys website.

Server vendors run the VMmark test under a set of guidelines and submit results to VMware for posting. It is my suspicion that vendors play leapfrog with VMmark by looking at exsiting VMmark results and only submitting their performance results when theirs are as good or better.

For instance, IBM submitted a benchmark for its 16-core System x3850 M2
running VMware ESX v3.5, which trumped the other published as of March 2008. IBM then published a press release to brag about the results, but within a few months, Dell submitted results three PowerEdge systems sporting better virtual machine (VM) performance than IBM, and Hewlett Packard (HP) beat them all out with its ProLiant DL585 G5 server results published August 5.

HP also sent out an email to press this week boasting their top 32-core results, but didn’t mention one minor detail; they are the only vendor with results in the 32-core category so far. Sure, they are number one. They are the only one.

System Administrator Bob Plankers sums this game up nicely in his blog with a post called “Why VMmark Sucks.”  Here is what Plankers had to say:

“Having a standard benchmark to measure virtual machine performance is useful. Customers will swoon over hardware vendors’ published results. Virtualization companies will complain that the benchmark is unfair. Then they’ll all get silent, start rigging the tests, scrape and cheat and skew the numbers so that their machines look the greatest, their hypervisor is the fastest. Along the way it’ll stop being about sheer performance and become performance per dollar. Then CapEx vs. OpEx. Watt per tile. Heat per VM. Who knows, except everybody will be the best at something, according to their own marketing department.”

In addition, the benchmark is a real pain to set up and run, and the ‘free’ VMmark software requires other expensive software to work. According to VMware’s website,
VMmark requires  licenses for the following software packages;

• Microsoft Windows Server 2003 Release 2 Enterprise Edition (32-bit)—thre 32-bit copies per tile (two for virtual machines and one for that tile’s client system), and one 64-bit copy per tile (for the Java server virtual machine)
• Microsoft Exchange Server 2003 Enterprise Edition
• SPECjbb2005 Benchmark
• SPECweb2005 Benchmark

Plankers said he won’t be wasting any time or money running VMmark. “Instead, I’ll be in meetings explaining to folks why we are maxed out at 30 VMs per server when the vendor says they’ll run 50. Or why we chose VMware over Xen, when Xen claims 100 on the same hardware. I’ll have to remember the line from the FAQ that says “that VMmark is neither a capacity planning tool nor a sizing tool.”

Which begs the question: if it isn’t for use in sizing or capacity planning, exactly what is it good for?”

VMware says the benchmark is good for users who are making hardware purchasing decisions.

“The intention [of VMmark] is that customers can look at the results and make decisions based on what they see. It isn’t just about the fastest server; it’s about making system comparisons; between blades and rackmounts or a two-core or four-core system. Someone can see how much more performance they get from upgrading to four core processors, for instance,” said Jennifer Anderson, the senior director of research and development at VMware.

This makes sense, but as Plankers said, users should beware of benchmark manipulation by vendors and know that the results do not reflect the same workloads that users will run in their own data center environments.

When server vendors introduce new blade servers these days, they often mention virtualization in the same breath, often touting the number of virtual machines (VMs) their hardware can support. But those numbers are hardly the result of scientific method.

For instance, San Diego, Calif.-based Verari Systems recently announced that its VMware ESX 3.5-certified VB1257 for BladeRack 2 XL supports up to twice as many VMs as competitive offerings (16). After speaking with Verari, I asked the competition — Sun Microsystems, Hewlett-Packard and Dell — how many VMs their blades can hypothetically support, and was given some big numbers.

But are these server vendors asking the right question? According to Anne Skamarock, a research director at Focus Consulting, the answer is no. Although vendors boast about the number of VMs their hardware supports, “it really is a silly way to look at it,” she said.

“The number of VMs supported depends on the workload. For CPU-intensive workloads, memory will also be a significant factor in performance,” Skamarock said. ”I have spoken with customers who are running 30 VMs per 8-core system and expect to increase that to 50 VMs per system.”

Skamarock said Virtual Desktop Infrastructure adds another twist. “The rule of thumb is six to eight virtual desktops per core, but again, memory will be a big issue here depending on the OS.”

According to preliminary data from SearchDataCenter.com’s 2008 Purchasing Intentions Survey, 61% of the respondents run less than 10 VMs per server, though 33% run 10 to 25, and a mere 5% run more than 25 VMs on a server.

Vendors make big VM support claims

According to VMware Inc.’s website, server consolidation ratios commonly exceed 10 virtual machines per physical processor; so presumably, a blade server with two CPUs, like Verari’s VB1257, should be able to support at least 20 VMs.

Within HP’s ProLiant blade server line, the ProLiant BL460c/465c and BL680c/BL685c would be a good choice for a virtual server platform, primarily because they offer a large memory footprint, which means more than 16 VMs per blade in both cases, plus more network expansion and storage performance, HP spokesman Eric Krueger said.

“Keep in mind of course the number of VMs always vary – the number could be higher or lower depending on the needs the application/VM — but based on the rule of thumb … the BL460c can support up to 16 VMs and the BL680c up to 32,” Krueger said.

Sun Microsystems Inc. claims its Sun Blade servers pack two and three times that many VMs. The Sun Blade X6250, which has up to eight cores with Intel Xeon processors, 64 GB RAM, 110 Gbps I/O and 800GB of internal storage, supports 36 VMs; the Sun Blade X6450, with two or four dual-core or quad-core Intel Xeon processors and up to 96 GB of memory, can support up to 42 VMs and the Sun Blade X8450 with 16 cores per module and 128 GB Memory, supports up to 48 VMs, according to Sun.

Dell was hesitant to name a number of VMs that its PowerEdge blade servers can support, because the number is dependent on a number of factors, like workload, memory, I/O. A spokesperson did say that “Dell has blades that support up to 66 loaded VMs. This is based on VMware’s VMmark benchmark test,” a spokesperson said. “This is an area where we are doing quite a bit of work, so stay tuned.”

So I’m wondering: Are VM support numbers a consideration when buying server hardware, or is it too subjective? Let us know what you think.