VMware: Virtual Reality

As you may already know, VMware recently announced an OEM agreement with Novell to redistribute SUSE Linux Enterprise Server (SLES) to eligible VMware vSphere customers. Just a few hours later (wow – that was quick), Microsoft published its take on the Microsoft Virtualization Team blog. Unfortunately, fast is not always a synonym of well-thought out. The arguments presented in the Microsoft blog not only miss the point about the announcement, but are so far off base that it makes one wonder whether Microsoft has learned anything about virtualization or is just trying to generate some headlines for headline’s sake.

In either case, given the level of marketing spin in the Microsoft blog post, I feel obliged to address the most blatant misinformation and set the record straight:

Myth #1: “Looks like VMWare finally determined that virtualization is a server OS feature. I’m sure we’ve said that once or twice over the years ;-)”

Our announcement is about providing SLES as a guest operating system (OS) and not as a hypervisor. Offering a more cost effective way to deploy SLES in VMware environments has nothing to do with the architecture of the hypervisor. Come on Microsoft – this is Virtualization 101 level stuff. VMware is committed to a hypervisor architecture that does not rely on a general purpose OS, unlike Hyper-V’s reliance on Windows, as it is a fundamentally better design that leads to higher reliability, robustness and security. This is why in our latest generation hypervisor architecture - VMware ESXi - we removed the console OS. Independent industry analysts agree that a slimmer hypervisor is the right approach – see “A Downside to Hyper-V”. We certainly don’t plan to reverse our direction, quite the opposite actually. As we publicly stated multiple times in the future ESXi will be the exclusive focus of VMware development efforts. Thanks to the ESXi hypervisor architecture, our customers won’t run the same risks they would have to face with Hyper-V.

Myth #2: “The vFolks in Palo Alto are further isolating themselves within the industry. Microsoft’s interop efforts have provided more choice and flexibility for customers, including our work with Novell.”

1. VMware vSphere supports 65 guest operating systems versus Hyper-V R2 supporting only 17
2. VMware vSphere supports more Microsoft operating systems than Microsoft Hyper-V R2 itself
3. VMware vSphere supports 6 times more Linux operating systems than Hyper-V R2

Who is isolating itself? Who provides more choice?  Let’s not forget that just a few months ago Bob Muglia, President of Microsoft’s Servers and Tools business unit, stated that the number #3 top competitor for his division in 2010 is Linux! And just this past week at TechEd 2010, Steve Ballmer listed “Open Source” as a top competitor for Microsoft. How is it possible that on one hand Microsoft touts “new interoperability” with Linux and on the other one wants to kill it? Something has got to give and I think I know which one it will be….

Myth #3: “As one of many examples of our work with open source communities, we’re [Microsoft] adding functionality to the Linux Integration Services for Hyper-V. In fact, we have an RC version of the Linux Integration Services, which support Linux virtual machines with up to 4 virtual CPUs. In fact, we’ll talk more of this on June 25 at Red Hat Summit.”

VMware has a track record of providing equal support to Windows and Linux operating systems. We have supported 4 virtual CPUs for Windows and Linux guest OSs since 2006 and added 8-way vSMP in 2009. The OEM agreement with Novell doesn’t change our commitment to guest operating system neutrality. Positioning the Hyper-V’s upcoming support of 4 virtual CPUs for a small subset of Linux operating systems as a big win only confirms that 1) Microsoft is failing to keep up with VMware, and that 2) Microsoft has treated Linux guests as second class citizens. Is it credible that this second-class status for Linux will somehow change given that Linux and Open Source are being classified as top competitors?

Myth #4: “This is a bad deal for customers as they’re getting locked into an inflexible offer. Check out the terms and conditions. [..] So be sure not to drop support or you’ll invalidate your license”

So, before customers had to purchase SnS for VMware vSphere and a SLES subscription for patches and updates. With this new VMware-Novell agreement, they only have to purchase VMware SnS. Help me understand how this is a bad deal for customers? Talking about “invalidating” licenses in the context of a Linux operating system doesn’t make much sense given how the Linux licensing model works. The SLES deal offered by VMware is about subscription to patches and updates and not about licenses. Applying the same logic to Microsoft Software Assurance, we should warn customers that Microsoft SA is a bad deal because it locks them in an inflexible offer that forces them to pay in order to get the next Windows upgrade.

Myth #5: “Last, the vFolks have no public cloud offering, like Windows Azure, like Amazon EC2. While we’re demoing and building capabilities so customers have a common and flexible application and management model across on-premises and cloud computing, they’re stitching together virtual appliances to fill the void.”

Microsoft clearly “forgets” about VMware’s 1,000+ vCloud partners and public infrastructure as-a-service solutions based on VMware technology like vCloud Express . Our objective is to enable a broad partner ecosystem of service providers that leverage VMware’s technology to offer cloud services. This will give customers freedom of choice. We also want to make sure that companies retain control of their applications and are not locked in to any one particular service. Virtual appliances are a key component of this strategy, because that’s ultimately where the application lives. Microsoft isn’t interested in virtual appliances, because it isn’t interested in enabling application portability among cloud provider. Ultimately Microsoft’s strategy with Azure it to have customers run applications on Microsoft operating systems using Microsoft databases in Microsoft datacenters…. looks like the mother of all lock-ins.

Is Microsoft suffering from Hyper-Desperation R2?

Such an incredibly off-base reaction is clear evidence that the VMware-Novell OEM agreement struck a nerve at Microsoft. Could it be a sign of Hyper-Desperation R2? After all, the events of the past 2 months must have been pretty hard on the nerves of the Microsoft Virtualization Team:

• On April 27^th and May 19^th , VMware announces new technology partnerships with two major cloud computing vendors, Salesforce.com (“Salesforce.com and VMware Form Strategic Alliance to Launch VMforce™, the World’s First Enterprise Java Cloud”) and Google (“VMware to Collaborate with Google on Cloud Computing”). This strategy offers far more choice to customers than Microsoft’s Azure-only approach
• Then, on May 26^th , Gartner publishes the 2010 x86 Server Virtualization Magic Quadrant, placing VMware in the “Leaders” quadrant, thereby demonstrating our clear leadership
• Finally, just last week at Microsoft TechEd 2010 (one of Microsoft’s biggest shows of the year), VMware vSphere wins the “Best of TechEd 2010” award in the virtualization category and the “Best of TechEd Attendee’s Pick” awards. It must have been unsettling for the Microsoft virtualization team to see attendees at their own conference vote for VMware vSphere

But all of this aside, at the end of the day, what really matters is that VMware continues to show strong execution in our mission of simplifying IT and providing customers a pragmatic path to the cloud. Our agreement with Novell is another great example of how we’re delivering on our mission.

VMware has been exhibiting here at Microsoft TechEd 2010 and lots of VMware customers, plus those just starting the virtualization journey, have come by to talk to us - lots of questions about how to virtualize applications like Microsoft Exchange, SQL, and Sharepoint on VMware vSphere, how to implement an effective disaster recovery plan for their virtual Windows environment, and how to leverage VMware View for their Windows 7 migrations. But also, a lot of folks come by simply to say hello and share how well their Windows environments run on vSphere. Thanks to those who came by our booth this week.

I must confess though, that even after talking to all these customers, I did not fully understand how many satisfied VMware customers are here at Microsoft TechEd 2010… here’s what I mean.

Wednesday night, the VMware team attended the Best of TechEd awards ceremony because VMware vSphere was a finalist in the Virtualization category. We won that award last year (Best of TechEd 2009) and were anxiously waiting to see if we would win again this year. We were very excited when the judges announced VMware as the Virtualization category winner for Best of TechEd 2010.

Soon after, the final award of the night came up – the Best of TechEd Attendee’s Pick Award. This award is 100% based on votes from the 10,000+ TechEd attendees (my guess, as I haven’t seen an official number) – no judges involved. Attendees got to choose from among all the IT products exhibited at Microsoft TechEd, not just the virtualization ones, basically everything in the world of Windows IT. Given this context, we were thrilled when VMware was announced as the winner of the Best of TechEd 2010 Attendee’s Pick Award! Wow, to be chosen by the attendees from among all the IT products at TechEd, was just the latest testament of the value that vSphere delivers to customers.

There were no acceptance speeches at the awards ceremony (which was probably a wise decision by the folks at Windows IT Pro), but some thank-you’s are definitely in order. To the 25,000+ VMware channel partners, thanks for educating and helping customers implement vSphere to solve their most pressing datacenter needs. To the 35,000+ VMware Certified Professionals (VCPs), thanks for managing all of the vSphere installations so they run smoothly, and for being the resident experts in your respective companies, sharing with everyone the value of VMware. To the 1,500+ VMware technology partners, thanks for working with our engineering teams to build tight, robust integrations between our respective products. And finally, thanks to the thousands of VMware R&D engineers who built VMware vSphere – great job team.

Sometimes, what one does not say, communicates a lot more than what one actually does say. That seemed to be the case last week when Microsoft released new collateral telling channel partners that they can achieve higher growth by selling Microsoft virtualization products alongside their existing VMware practices.

Unfortunately for Microsoft, the blogosphere heard a different message:

• “Microsoft admits that VMware has been the only choice for partners so far” (virtualization.info, May 24, 2010)
• “Bottom line: Despite the heaping helpings of hype, Hyper-V is not capturing the hearts, minds, or pocketbooks of Microsoft's sales partners at the expense of VMware.” (Virtualization Review, May 24, 2010)

Wow – talk about an unintentional message. Instead of Microsoft’s intended message, people picked up on Microsoft’s change in tactics from “sell Hyper-V because it can go head-to-head with VMware” to “why don’t you try selling Hyper-V in addition to VMware – we promise it will make you more money” (my paraphrase). To quote the Virtualization Review article, “While this is not tantamount to raising a white flag, it is a stark, public acknowledgement of VMware's deeply embedded pre-eminence.” Thanks Microsoft for reaffirming what VMware partners and customers have known all along.

Microsoft Tool: Partners Get Better Margins with VMware

But, unintentional messages aside, is there truth behind the claim that Microsoft virtualization products can make partners more money? To answer this question, I spent some time playing with Microsoft’s Excel model (the heart of the new Microsoft collateral) that compares the revenue and margins from selling VMware versus Microsoft.

First of all, Microsoft’s model states that partners generally make better margins by selling VMware (default assumptions in the Microsoft model).

Microsoft Tool Assumes Up to a 2x Win Rate Advantage over VMware – Quite Unrealistic

But if that’s the case, what is Microsoft’s argument for how partners can make more money selling Microsoft virtualization over VMware? As Steve Kaplan points out on his blog, the two biggest assumptions in the Microsoft’s model are 1) a higher win rate (i.e. closing more deals) when leading with a Microsoft virtualization offering, and 2) “upsell capture” where customers purchase more software/ hardware/services with the money they supposedly save with a Microsoft offering. (There are a bunch of other default assumptions that disadvantage VMware, which I list at the end of this blog, but higher win rate and upsell capture have the biggest impact on the model’s output.)

So what does the tool assume regarding win rates? The default ‘mature market’ assumption is that leading with Microsoft virtualization offerings will win 33% more Large Enterprise deals and 100% more Mid-market deals (2x) compared to if the partner leads with VMware offerings, resulting in partners making more margin with Microsoft virtualization. Those are pretty bold, and frankly, unsubstantiated assumptions given VMware’s dominant position in Large Enterprises (96% of the Fortune 1000 are VMware customers), VMware’s strong traction with our mid-market offerings (Acceleration Kits, Essentials Editions), and the fact that customers get a more reliable, more complete, and lower cost-per-application solution with VMware.

Just how sensitive is the Microsoft tool to the win-rate assumption? Let’s find out by making two simple adjustments to win-rate. Let’s give Microsoft the benefit of the doubt with a higher win rate in mid-market (33% more deals), put us on parity in the Large Enterprise (same win rate), and leave in place the model’s assumption that customers will buy more hardware/software/services because of the money they save with Microsoft (an assumption that Steven Kaplan challenges in his blog, but that’s another discussion). Now the VMware practice makes $1,021,000 of margin at the end of Year 3 while the Microsoft practice makes $948,000 of margin – so the VMware practice now makes more money, even with a 33% higher win-rate assumption for Microsoft in mid-market!

Microsoft Tool’s Default Assumptions Are Just Not Credible

This simple exercise is revealing because it shows what partners would have to believe in order for Microsoft virtualization to make them more money, essentially, that customers would have to overwhelmingly choose Microsoft virtualization products over VMware by an incredibly higher, unsubstantiated rate, in spite of the fact that VMware delivers far more compelling benefits to customers:

• a more reliable and robust product,
• a more complete platform for virtualizing applications,
• a more comprehensive set of virtualization management solutions,
• more customer choice with broader OS, application, and hardware support,
• a more proven solution as evidenced by customer adoption and return on investment, and
• a lower cost-per-application - a key component of total cost of ownership for a virtual environment.

Now, I realize that there are many VARs and consultants out there that have built very successful, profitable businesses selling Microsoft products, so my point is not about any channel partner’s existing Microsoft practices. But I do take issue with Microsoft’s claim that selling Microsoft virtualization is more profitable than selling VMware based on their “model” – Microsoft’s assumptions to make their case are just not credible.

VMware Channel Partner Ecosystem: 25,000+ Partners and Growing

To close, here are some facts about the ecosystem of VMware partners:

• VMware has over 25,000 channel partners worldwide and that number is growing.
• For every $20k of VMware licenses sold, partners sell another $225k of hardware/software/ services. (VMware research, 2008)
• 75%+ of VMware’s revenue comes from our channel partners.

As you can see, VMware and its channel partners have a strong win-win relationship.

If you are a VAR or consultant and currently not a VMware channel partner, we invite you to check out the VMware Partner Network website to get information about how to become one. Let us help you grow your business by selling VMware while delivering the most trusted, most complete, lowest cost-per-application solution to your customer.

A Few Other Default Assumptions in the Microsoft Tool to be Aware of

• Default assumption charges VMware projects a 10% premium on hourly labor.
• Default assumption is that VMware projects take 10% more engineering hours to complete.
• Default assumption results in partners completing almost twice the number of Microsoft projects as VMware projects in the same period of time without adding any additional headcount. Not realistic at all.
• Default assumption appears to use an inconsistent mix of list price and channel price for Microsoft and VMware products. (I could not figure out a consistent methodology to the numbers. If you figure it out, make a comment below.)
• Default assumption adds the cost of vSphere Enterprise Plus licenses in one scenario without explaining why VMware’s highest-end SKU was required, and there’s no acknowledgement that Enterprise Plus delivers far more functionality than the Microsoft offering.
• Default assumption adds the cost of VMware Site Recovery Manager (SRM) licenses, our disaster recovery product, to another scenario without adding anything to the Microsoft stack, in spite of the fact that Microsoft has nothing comparable to SRM.

One of the fun parts of my job is meeting with customers when they visit VMware’s Briefing Center here at our headquarters in Palo Alto, California – in the heart of Silicon Valley. The customers range from those who have used VMware for years and are now working towards the next step, towards a Private Cloud, as well as, those who are newer to virtualization and have begun server consolidation but are also starting to realize how a virtual datacenter can deliver so much more than just CapEx savings – open conversations to figure out how VMware can help the customer (further) reduce complexity, lower costs, and boost business agility.

In these customer meetings, a question that comes up from time to time is “We know you guys are the clear leaders today. How do you intend to maintain your leadership position moving forward?” A fair question, upon which I walk them through our overall business strategy to enable IT-as-a-Service through cloud computing, product roadmaps, and our competitive differentiation (most of which are captured on our Why Choose VMware and Advantages of VMware Virtualization web pages).

But if you read between the lines, their question is also about an attitude check – one that is often posed to industry leaders. It is a check to make sure that VMware’s success has not somehow made us complacent in regards to changing market dynamics or offerings from other vendors. Basically they are asking, “Are you guys resting on your laurels?” And the answer is “Absolutely not!”

Speaking from my experience working at VMware, I see as much (and very likely, more) ambition, drive, and energy now, as when I joined VMware three plus years ago. A survey of some of our announcements over the past year serves as good evidence of our aggressive drive to enabling IT-as-a-Service:

• Release of vSphere 4, the latest major version of our flagship product that is the foundational virtualization platform for IT-as-a-Service, garnering a large stack of industry awards and recognitions over the past 12 months;
• Release of View 4 with its new PCoIP protocol, the industry’s only purpose-built desktop virtualization solution for delivering desktop-as-a-service;
• Acquisitions of SpringSource, Zimbra, and other companies to broaden the overall solution stack that VMware will offer to customers – from the platform to developer frameworks to cloud-ready applications;
• Significant advancements in our vCloud initiative, such as having more than 1000 service provider partners in the VMware Virtualized service, including partners like AT&T, SAVVIS, Terremark, and Verizon Business.
• Formation of vmforce.com, a strategic alliance between VMware and salesforce.com, to deliver a trusted cloud for enterprise Java developers;
• Active collaboration with industry bodies like the DMTF to enable cloud computing standards and interfaces to maximize customer choice and flexibility in this next emerging era of computing;
• And the list can go on…

Attitude is important. As our CEO Paul Maritz likes to remind us, “Beware the good times.” There is wisdom in that exhortation and echoes the sentiment from former Intel Chairman and CEO Andy Grove’s famous quote, “Only the paranoid survive.” Both men know what it takes to be and maintain leadership in the ever changing technology industry. From my view, there’s a healthy level of “paranoia” at VMware these days – the right level to continue (and extend) our leadership position.

Last week VMware announced very positive business results for the first quarter 2010, beating analyst expectations by a considerable margin. Needless to say, we are thrilled to see how customers continue to turn to VMware to reduce complexity, dramatically lower costs, and increase business agility.

In the earnings call, our execs also described how we saw strong momentum in our vSphere bundles for small and medium businesses (vSphere Essentials, vSphere Essentials Plus and vSphere Acceleration Kits). For some people, hearing this may be a bit of a surprise as other virtualization vendors have been trying to create the misconception that VMware’s virtualization is too expensive for small businesses. Microsoft, for example, claims that Hyper-V and System Center are 3x to 12x less expensive than the VMware’s Essentials bundles. Ultimately, nothing better than positive business results demonstrates how such claims are a misrepresentation of reality, but just in case, here is some simple math that proves it.

In the table below we compare the cost of software licenses and support to deploy a virtualization solution with centralized management and the ability to quick recovery from hardware failures. The size of the deployment is 3 virtualization hosts (2 CPUs per host) running 15 VMs total.

As you can see from the table, VMware Essentials Plus is actually about $800 cheaper than a comparable offering from Microsoft even when only considering licensing cost and assuming equal consolidation ratios. While we believe that cost per application is a better approach to determine cost of acquisition for large and small/medium businesses, we wanted to debunk our competitor’s claim using their preferred methodology – upfront software license costs.

Price is certainly an important element of consideration for a small business; however it is not the only advantage that explains the momentum of the VMware Essentials bundles. Reliability, ease-of-use, unlimited technical support calls, and rapid recovery from unplanned hardware failure are also other very important ones. In the last couple of blogs (“VMware ESX to ESXi Upgrade Center – Check it Out!”, “Jumpstart a free virtual environment in a few clicks with VMware GO and ESXi”) we have talked about how and why VMware ESXi is a more reliable, robust and easy-to-use virtualization platform. VMware’s solutions for SMBs are based on the same core technology that the largest companies in the world use in their production environments -- there just is not a more proven x86 virtualization platform out there. Why settle for less when you can afford to get the best? I guess many people have already figured this out.

In my next blog, I will discuss in greater detail the advantages that vSphere Essentials Plus offers for recovery and protection from hardware failures.

In our last blog post, we discussed what motivated VMware to create ESXi and how its unique streamlined architecture, with no dependence on a general purpose operating system (Windows or Linux), makes it a more reliable, secure and efficient virtualization platform. These benefits are fundamental advantages of ESXi over other hypervisors, however there’s more. In fact, ESXi is also the ideal solution to quickly jumpstart a free virtual environment without a lot of pre-existing virtualization expertise.

The first key ingredient for jumpstarting a free virtual environment is a simple installation and set up process. The video linked below that the guys at GoVirtual.tv put together shows how installing ESXi is very straightforward and can be completed in a few minutes with basic system administration skills.

While a simple installation process for a hypervisor is certainly important, it represents only the initial step towards the ultimate goal of migrating applications from physical servers to virtual machines and getting them up and running as rapidly and smoothly as possible. For most large companies these are now well established processes of datacenter operations, however for first time users or resource constrained small business they could still represent a roadblock towards virtualizing apps. To help remove this roadblock, in January 2010, VMware launched VMware GO - a free web service for SMBs to help them virtualize their physical servers on ESXi and have the new virtual machines up and running with just a few mouse clicks.

VMware GO eliminates the few manual operations still present in ESXi’s installation (like verifying the hardware compatibility list or burning a CD with ESXi’s image), provides everything needed to deploy VMs on ESXi hosts, and offers basic management and reporting capabilities. Here is a list of what VMware GO can do:

• automatically check hardware for ESXi compatibility
• download ESXi and burn installation CD
• verify configuration of new ESXi installation
• apply NTP settings
• convert physical servers to virtual machines
• create, delete and resize VMs
• download and deploy virtual appliances from VMware’s Virtual Appliance Marketplace
• open a remote console to a VM
• scan ESXi hosts and VMs for patches
• generate reports: ESXi host inventory, VM inventory, patch status

You can see VMware GO in action in this great video that Dave Davis at TRAINSIGNAL recorded (Dave’s video is based on the beta version of VMware GO, nevertheless it is a very good overview of actual capabilities).

The video clearly shows how VMware GO perfectly complements the already simple installation process of ESXi, making it extremely fast not only to set up the hypervisor, but also to deploy virtual machines and perform basic centralized management tasks – all for free! Keeping in mind the unique benefits of ESXi’s architecture discussed in our previous blog post, it is easy to see how by providing greater advantages in terms of reliability and ease of use than even Microsoft Hyper-V or Citrix XenServer VMware GO + ESXi is a better solution for first time users and budget constrained small businesses.

If you have not had a chance, I would definitely recommend you check out the recently launched “VMware ESX to ESXi Upgrade Center.” We launched this upgrade center to help customers and partners get comfortable with ESXi because we’ve stated on numerous occasions that in the future, ESXi will be the exclusive focus of VMware development efforts.

Why have we made this decision to move towards the ESXi architecture? Simply put, we wanted to improve hypervisor management in regards to security, deployment and configuration, and ongoing administration. You can read more about all the benefits at the Upgrade Center or ESXi product page, but as a quick snapshot, take a look at this graph of how ESXi reduces the number of patches compared to traditional ESX (over the same timeframe).

It just goes to show you what removing the dependency on a general purpose OS in the virtualization layer can do for you. Goodbye patches for general purpose operating system code that wasn’t required for core virtualization functionality. (Although the console OS in traditional ESX serves a vastly different purpose than the Parent Partition in Hyper-V or Domain 0 in Xen, but that’s a whole separate discussion.)

As expected, not everyone agrees with our approach of reducing the role of a general purpose operating system in the virtualization layer. Just take a look at a Microsoft marketing brochure that proudly highlights how their virtualization layer is part of general purpose Windows. But it is very interesting how, once you talk to knowledgeable people outside of the Hyper-V marketing team, you find resoundingly consistent agreement that a smaller code base is the way to go for protecting your mission-critical virtualization layer.

• From a leading virtualization security analyst’s blog (Feb 2010):
  “The lesson from all of this is that thinner is better from a security perspective and I’d argue that the x86 virtualization platforms that we are installing (ESX, Xen, Hyper-V and so on) are the most important x86 platforms in our data centers. That means patching this layer is paramount. With Hyper-V’s parent partition that means closely keeping an eye on Microsoft’s vulnerability announcements to see if it is affected.” (emphasis added)
• From an article quoting another leading virtualization analyst (Nov 2009):
  “Problem two is a bigger, architectural problem that he was told about by R2 beta users. As he explains it, in a Hyper-V environment, every physical host has a copy of Windows that is used as the parent OS. It manages the I/O drivers and is home to any management agents that are installed. ‘If I want to use PowerShell, I'm also using the parent OS for that,’ he declares, ‘so what you end up with is one big, fat, single point of failure.’” (emphasis added)
• Finally, let’s look at what a Director in the Microsoft Azure team said was a key design principle for Azure:
  “[Design Principle #2] Small footprint: any features not applicable to our specific cloud scenarios are removed. This guarantees that we do not have to worry about updating or fixing unnecessary code, meaning less churning or required reboots for the host.” (emphasis added)

Pretty good independent confirmation of our approach with ESXi.

So give ESXi a try if you’ve never done so. If you already own vSphere licenses, you have the option of deploying any of those licenses as either ESXi or as traditional ESX. If you don’t have vSphere, download ESXi for free.

While browsing through the Microsoft Virtualization website, I stumbled across this table included in the Cost Saving section that presents cost and feature checklist comparison between Hyper-V/System Center with few vSphere editions.

While Microsoft’s spin on the theoretical cost advantage of Hyper-V/System Center over vSphere isn’t surprising (I am not going to address it here, since we have already shown how it doesn’t hold water), the checklist comparison struck me as having a few factual errors and misrepresentations of actual product capabilities which I think are worth pointing out:

• vSMP Support – Microsoft’s support for vSMP is actually much more limited than the table shows. Hyper-V R2 supports 4-way vSMP only in VMs running Windows Server 2008 and Windows 7. For Windows Server 2003 VMs, Hyper-V R2 supports up to 2-way vSMP and for Linux (SUSE /RHEL) VMs just single virtual CPU. vSphere, on the other hand, supports up to 4-way vSMP with Standard, Advanced and Enterprise Editions and 8-way vSMP with Enterprise Plus edition on any vSphere supported guest OS (over 50 versions).

• HA/Clustering – The table incorrectly shows that vSphere Standard does not include HA/Clustering, when in reality it does. Microsoft seems also very generous with Hyper-V by implying it provides equal HA capabilities as vSphere. Unlike vSphere, for example, Hyper-V R2 does not provide VM restart prioritization, which means that there is no easy way for admins to make sure that critical VMs are being restarted first. Incidentally, the lack of VM restart prioritization is one the reasons why Burton Group stated that Hyper-V R2 is not an enterprise production-ready solution. In addition because Hyper-V R2 lacks memory overcommit (a feature that is totally missing from Microsoft’s checklist), it can restart VMs only if the failover server has enough spare memory capacity to accommodate the VMs of the failed host.

• Hot add – Microsoft gives Hyper-V R2 a check on Hot Add and then below the checkmark specifies “Storage” to indicate Hyper-V supports only Hot Add of a VM’s virtual disk capacity. vSphere gets a checkmark too, but what the table doesn’t tell is that it not only provides Hot Add of a VM’s virtual disk capacity, but also of virtual memory and CPU

• Storage VMotion – This checkmark is funny to say the least. If you don’t know what the word “quick” means in Microsoft’s marketing jargon (and believe me I have heard illuminating translations of the term from Microsoft’s own employees), you’d think that Microsoft has a fast Storage VMotion (possibly faster than VMware’s). The reality is that even just talking about Storage VMotion in Hyper-V’s case doesn’t make sense, because Microsoft’s Quick Storage Migration, just like Quick Migration for VMs, cannot migrate VM virtual disks without downtime. VMware Storage VMotion, on the other hand, can migrate virtual disks without any application downtime.

• DRS/PRO – Even now that Hyper-V has live migration, positioning PRO as a DRS-equivalent isn’t accurate. PRO is a fundamentally less usable and more complex solution for resource balancing. Unlike DRS, which can be configured from vCenter in a matter of few clicks, PRO Tips requires both System Center Virtual Machine Manager (SCVMM) and Operations Manager (SCOM). As Microsoft TechNet shows, SCOM is a very complex product that consumes a considerable amount of servers and databases that - opposite to what Microsoft wants people to believe - are neither free nor included in the cost of SMSD licenses. In addition to being hard to set up, PRO is dependent on software packages (PRO Packs) that each hardware vendor creates for its own products Last but not least, PRO lacks a global view of the resources of a group of servers (like DRS does with Resource Pools) and consequently it cannot optimize resource allocation across a cluster, but only react to the local conditions of a certain workload.

• vNetwork/Host Profiles – in my opinion, this line wins the Oscar for best checkmark in a “mis-leading” role. First, Microsoft drops the words “Distributed Switch” from VMware’s vNetwork Distributed Switch (vDS) making it look like a generic virtual networking feature. Then, it gives Hyper-V R2 a check for the vNetwork/Host Profiles combination implying that System Center also provides the same functionality as VMware Host Profiles when in reality the only way it could would be through extensive development of custom scripts and customization of SC Configuration Manager (should we include the extra cost to the System Center price at the top of the table?)

While there is more that could be said about this table from Microsoft, this already shows how easy it is for Hyper-V to pass Microsoft’s checkmark exam. . This isn’t something new, though. Looking through the Virtual Reality archives, I found a 2 year old post (“Can I have the check, please?”) by a former VMware SE now with Microsoft on this same checkmark issue. I guess it is true that old habits die hard.

If you’ve been to any Microsoft virtualization session over the past year, you’ve seen Microsoft comparing VMware vs. Microsoft costs to deploy 5 virtualization hosts and how VMware is supposedly much more expensive. We’ve already blogged in great detailabout why Microsoft’s approach makes little sense in a virtual environment (because it totally ignores VM density, which is critical for accurate cost calculations in a virtualized datacenter) and how the claim itself is not credible.

But there is another point that needs to be considered – the cost of support. In these Microsoft cost comparisons, Microsoft likes to compare the cost of VMware Production SnS (25% of license price per year, formerly known as Platinum SnS) to that of Microsoft Software Assurance (25% of license price per year). VMware Production SnSincludes subscription and 24x7, unlimited support to give our customers the peace of mind of knowing they can get support when they need it. Microsoft SA, on the other hand, includes subscription, but only one support incident for every $20,000 of server/CALs licenses purchased.

So what does a customer do once they’ve burned through the support incidents “included” with SA? Well, they have to buy more support from Microsoft of course. Option 1 is to purchase more per-incident support at $259 per incident(business hours only) or $515 per incident(after hours support). Option 2, if they are a larger enterprise, is to buy Microsoft Premier Support, which costs $210/hour (purchased in blocks of pre-paid hours). Microsoft’s cost comparisons to VMware leave out these additional support costs.

Now, your final support cost with Microsoft will depend on your environment and support requirements. But the next time you hear about or decide to do a cost comparison, remember to factor in VM density and the additional Microsoft support costs beyond SA.

It’s now widely accepted that virtualization is one of those industry-wide “techtonic” shifts that, like previous shifts to client-server architecture or to the Internet, mark the beginning of a major transformation in how datacenters are architected and run. However, the virtualization transformation doesn’t stop there. The recurring theme that we hear from customers and partners is that the transformational effect of virtualization goes far beyond technology to directly impacting how IT looks at the economics and cost structure of the datacenter. At the core of this economic shift is the fact that in the virtualized datacenter, applications run on shared compute resources, not on dedicated ones like in the old physical datacenter. Customers tell us that in the new reality of the virtual datacenter the main variable driving planning and budgeting activities isn’t the number of physical servers any more, but the number of virtual machines (i.e. applications). In our experience, we have found that for this reason in the context of a fully or even partially virtualized datacenter application-based or VM-based models do a more accurate job of quantifying cost because they inherently factor in the economies of scale of multiple applications running on a shared infrastructure. It is very intuitive to recognize that, everything else equal, a higher the number of VMs per virtualization host, aka. VM density, corresponds to a higher infrastructure utilization rate and a lower average cost of running an application.

To help companies see how to implement a per-VM cost model for their datacenter, we asked IDC to work with one of our customers – Landmark Healthcare – and document how they are using an application-based cost model to track the cost structure evolution of their datacenter in their journey to the fully virtualized datacenter and the private cloud. The findings were documented in the IDC White Paper sponsored by VMware, The Economics of Virtualization: Moving Toward an Application – Based Cost Model.[1]

"Very clearly the existing process for the datacenter are undergoing a dramatic transformation. IDC believes that we are entering a new business cycle for IT and that virtualization will remain the foundation for enabling and driving a new set of economics for the datacenter. As customers increasingly deploy virtual environments they find themselves re-evaluating their traditional procurement and sourcing models. We continue to see cost per application as a more valid metric for measuring datacenter efficiency that will challenge existing server-based cost models." – Michelle Bailey, Research VP, Enterprise Platforms and Datacenter Trends

Landmark Healthcare – IDC Case Study Summary

Landmark Healthcare is an insurance provider for chiropractic medicine, currently generating $20+ millions in revenues with about 110 employees. The company is transitioning its business from that of a pure-play insurance company to one that specializes in claims processing and claims analytics for other health care providers. In 2008 Landmark decided to virtualize about half of their business applications in an effort to consolidate their server install base and contain their expected growth. Like many others, Landmark rapidly realized that the benefits of a virtual infrastructure could be extended well beyond the initial capital savings into areas like improved business continuity, faster application deployment, simpler management and higher productivity. Coming from the success of their initial virtualization phase and seeing the tremendous additional value they could get, Landmark now plans to virtualize all the remaining business applications and to invest in an iSCSI storage array to leverage business continuity features like VMware HA, VMotion, and Fault Tolerance. Once this second virtualization phase is completed in 2010, Landmark’s datacenter will be 100% virtual and will have enough extra capacity to support the expected applications growth for the following two years. In this time frame, Landmark expects it will only have to continue expanding its disk storage capacity according to the typical needs of its applications.

Landmark Healthcare – cost per application analysis

Working with Landmark, IDC built a detailed view of the company’s annualized cost per application at each stage, from a) fully physical to b) fully virtualized to c) planning for future needs. The chart below - taken from the IDC white paper – shows a summary of Landmark’s past, present, and projected cost-per-application and VM density (i.e. number of VMs per virtualization host).

As the chart shows, Landmark will reduce the average annualized cost of running its applications from $1,514 before virtualization (2008) to $643 in a fully virtualized datacenter (2009) despite a sizable investment in a new SAN. While this may sound like a big reduction, it is actually typical for VMware customers, who can achieve higher economies of scale by driving up hardware utilization without compromising application performance, reliability or availability. It is also important to note that the chart only shows the “hard” capital cost savings for Landmark’s infrastructure. It does not include a quantitative measure of the other benefits of virtualizing, such as increased admin productivity, shorter recovery time in case of system failure, and faster IT response time to business needs.

VMware’s technology has been instrumental for us to reliably virtualize our apps while maximizing the utilization rate of our hardware. As our datacenter becomes 100% virtual, we will lower the cost of running our apps by over 60% even despite a sizable investment in a new iSCSI SAN. At the same time we will improve availability, improve business continuity and enable our business owners to achieve a faster time to market for our services. The ability to drive up consolidation ratios with vSphere will curb the need for additional infrastructure investment. In fact, we expect that in the near future we will be able to support the forecasted growth for our business without the need for additional servers, simply by expanding our storage capacity to accommodate the natural demands of our business applications – Ron Davis, Senior Network Engineer, Landmark Healthcare

It is easy to predict that with the trend towards fully virtualized datacenters, private clouds, and hybrid clouds, application-based cost models will become a standard practice. Read about the details of the Landmark Healthcare case in the IDC white paper. In addition to the detailed quantitative analysis, the white paper also provides information on virtualization industry trends as well as practical advices on leveraging vSphere to maximize hardware utilization.

What’s your take? How are you measuring cost in your virtualized datacenter?

[1] IDC White Paper sponsored by VMware, The Economics of a Virtualized Datacenter: Moving Toward an Application- Based Cost Model, Doc # 220766, November 2009