Peter Christy

How Microsoft Lost the Virtualization War (when cheaper is wrong ...)

Microsoft didn't have to lose data center virtualization to VMware. It wasn't a question of inferior technology or product functionality. They lost because their business model was wrong.

Microsoft's Server and Tool Business (STB) has done extremely well by identifying markets where complex software functionality can be engineered for simple installation and use, and then brought to market at a disruptively low price point. Time after time this model worked: Windows Server, Systems Center, Exchange and SharePoint. Not surprisingly, when it came to virtualization, this was the business model: offer lots of functionality at a disruptively low-price. Why did it fail so badly in this case?

It failed because virtualization was different. Virtualizing a set of servers isn't like upgrading the operating system or Office. Virtualization enables significantly improved economics but along the way it requires significant changes in how the data center infrastructure is managed and how the data center is organized. Therefore it isn't surprising that customers didn't just jump into virtualization. Instead they brought in consultants to help them plan and implement, and along the way created a new channel focused on the delivery of virtualization solutions.

VMware had an "expensive" solution (although still a very smart investment considered by itself). Selling the VMware solution made a lot more sense for the channel because they made a lot more money doing it, and at beginning the VMware solution was domonstrably "better" than the Microsoft solution (more advanced functionality). So if a channel partner could make a lot more money selling VMware guess what they did?

As the Hyper-V solution caught up (and with System Center in many ways surpassed VMware) guess what happened? More or less nothing, because the channel had no incentive to switch brands. Yes, VMware has an "expensive" solution, and customers would like a "cheaper" solution, but that's not enough given the partner economics.

I finally understand what people were trying to say when they noted that Paul Maritz knew the Microsoft play book ("he had written it").  The genius wasn't just to understand your big competitor well; it was to jujitsu Microsoft (use their own weight against them).  The bigger question is why Microsoft didn't see it coming. The ultimate loss is enormous. Windows Server has done extremely well in the data center. System Center is remarkable management system. But now flanked by VMware on both sides it's going to be challenging to get out of the pincer.  Ballmer has changed the STB management. Microsoft is a smart and powerful company. Next move is necessarily theirs.

VMware VSphere 5: So What?

Last week VMware introduced vSphere 5 with a very professional, Web-based launch (best Paul Maritz talk I've ever heard). The reaction has been minimal... Why?  I think there are two reasons:

1. VMware doesn't really have any competition any more.
2. The adoption of Cloud solutions in the Enterprise is limited more by the rate at which the customers can evolve than by the state of the product offering (VMware is nicely ahead of the customers).

Two years ago, my thesis was that Microsoft would catch up with VMware in terms of virtualization functionality, and then bring the significant assets of System Center to bear competitively. But that just hasn't happened. Microsoft talks proudly of their share gain in hypervisors, but my friends at IDC say that really isn't gain against VMware in the data center. System Center is still a fine offering but Microsoft has nothing like vSphere as a platform offering on which to build private clouds (as you know, where I think the business is today). Microsoft is sandwiched between a VMware hypervisor and VMware service management. If that flanking continues, I don't Microsoft will thrive. 

This week both VMware and Microsoft announced earnings. VMware announced a very healthy 40% Y-o-Y growth rate. It's harder to parse out exactly what System Center revenue growth was but my best guess is that it's something in the 15+% range, not bad, but certainly not keeping up with VMW. A couple of years ago Microsoft argued that VMware and System Center were the only parts of the systems management segment that were growing, and that each was growing about the same rate. As you can see from the chart before, for a while VMware wasn't really growing license revenues: the overall growth came from a increasing service line with a growing installed base. But license revenue growth is back up.

 
Maybe there is another way to view the data, but it looks to me like VMware has sort of won this battle over the last year.

What's a Cloud? (Simon Crosby is a big, fat idiot)

OK, I admit that's cheap sensationalism. Simon Crosby is not "big" (tall, yes) and he is anything but fat or stupid. But I think that he gets a couple of things seriously wrong when he defines to Cloud to only apply to services you can rent by the hour, and when he rejects all of the Enterprise IT efforts as "Cloud" relevant because it is too labor intensive.

Cloud is an very relevant topic because of the dramatic and ongoing change in data center technology and operation. This acceleration is unquestionably catalyzed by production quality virtualization but the change isn't about virtualization; it's about automated "on demand" and "elastic" operations and how they accelerate the practicality of business models like service-oriented IT.

Simon's exclusion of the Enterprise activity (EMC's "Road to the Private Cloud") is wrong simply because that is where the majority of IT Cloud spend is occurring, and because it meets our definition precisely. Simon is absolutely right to point out that enterprise IT cost structures are very different from Amazon or Google cost structures. Today CIO's yearn for a day when the cost of the hardware is a primary issue but today it's not. Private Cloud computing is a pathway toward a cost structure that is much leaner in labor, but getting to that point will take many years. What Randy Mott did at HP ( Randy Mott blog entry ). It's a painful process because it requires decommissioning old and redundant applications, changing business processes and retraining people, but it's all an important part of getting to service oriented IT. Don't interpret a very different beginning point as enterprise IT not having the same aspirations or not finding great value in Cloud systems.

Simon says that it's not Cloud computing if you don't buy it as a service. When AWS first got wide attention, more than one guru said that no one would do private computing again because AWS was so much cheaper. Now with the likes of Netflix, EBay (and even amazon.com) implementing hybrid public and private strategies, the real economics are becoming more visible and common sense is prevailing. AWS is a lot more expensive than what EBay does internally, but if you have a highly peaky workload, the economics of a hybrid scheme still make sense, even if AWS cost 4X your internal costs.

Here's the critical point that isn't broadly understood yet: normal enterprise workloads aren't highly peaky, and if you put enough of them together in a large, shared data center (and movethe peaky dev/test and marketing loads out to the Cloud where they rightfully belong), enterprise IT demand is highly predictable and remarkably constant. So an automated, on-demand, large enterprise data center can run at much higher utilization than can AWS (because AWS is providing peak capacity for many), and it doesn't have to make a profit. Enterprise Private Clouds can and will be as automated and nearly as labor efficient as AWS (at the infrastructure layer) and it will be significantly cheaper for enterprises to run their own private Cloud than to outsource to Amazon, not to mention the clear performance and governance benefits.

It his previous Citrix role Simon had many reasons to argue that what VMware is doing isn't really Cloud computing but if you look at how enterprise IT is spending money and what their goals are, it would be a tremendous mistake to exclude this from the Cloud discussion.

Why Clouds Exist

At times it can be hard to understand what part of the Cloud discussion has substance and what is just marketing fluff. What I want to do here is argue that over the last ten years there as been a "perfect storm" of events that collectively led to today's Cloud world. Clouds are not just yesterday's time sharing or ASP offerings with fresh marketing lipstick. Here's the story.

Consider all that has happened over the period time since the Dot Com bust:

• The Intel architecture "won."  If you were planning a Dot Com venture your funders would probably tell you to use an Oracle database on Sun Solaris. In 2000, many businesses had strategic investments in the IBM Power Architecture (or HP's proprietary architectures). Today there is widespread agreement that Intel's X86 is good enough. Intel won the microprocessor battle. Without that architecture convergence there would be no Cloud.
• The network got a lot better. Broadband access wasn't common ten years ago. It is now. Without modern connectivity and bandwidth there would be no Cloud.
• Wireless networks expanded. Wireless build-out during the period was remarkable on a global basis bringing developing areas immediately into the networked world, bypassing entirely the wired phase.
• Mobile devices became powerful and smart. With Apple's iPhone deal with AT&T, mobile applications escaped the control of the mobile carriers. The evolution in wireless and mobile devices (and the rapid rise of consumerism -- employee chosen devices) has really changed the enterprise application and design model. This transition is an ideal time to also transition the front-end parts of the application to a Cloud implementation.
• We consolidated server applications into data centers. During the Dot Com boom, businesses spent profligately for IT, fearing that otherwise they would be "Amazoned." When the bust came there was broad retrenchment in IT spending (it's a cost center, remember). One of the easiest ways to reduce IT cost is to consolidate diverse applications into data centers where the administration and support can be local and shared. This cost-driven consolidation was an important precursor to today's Cloud world.
• Google demonstrated the promise of very large scale commodity computing. Ten years ago if you suggested building "mission critical" applications on anything other than "enterprise class" servers and storage you would be laughed out of the room. Google changed the game entirely. They showed that the most powerful systems in the world could be built from high-volume, commodity parts, and operated with high availability at remarkably low cost. In part the drive to the Cloud is motivated by the radically improved cost structure clearly visible downstream.
• Intel adopted multicore microprocessor design.  For years Intel converted semiconductor design and process improvements (the train under "Moore's Law") into faster CPU's by making the CPU's more and more complex.  In the end complexity yields diminishing returns. Early in the decade progress looked threatened: new processors got the same task done more quickly, but required more energy in the process. Finally Intel followed AMD and adopted a multicore strategy in which technology improvements are used to put more independent processors on the chip (more "cores") rather than to build more complex processors. It was a risky decision. It wasn't clear how many cores a desktop PC could use productively and the same problem existed in the data center where Windows servers were commonly built to each run on an separate server. The adoption of multicore designs enables a rapid increase the power of a data center of a given size and with a given power budget.
• VMware created commercial quality X86 virtualization and server consolidation occurred. Virtualization abstracts a workload from the specific details of the hardware on which it runs and enabled the shared use of a virtualized server by many X86 workloads with no requirement that they all ran the same "stack."

Virtualization is the magic lubrication that brings it all together: we consolidated applications into data centers; the victory of the X86 architecture and virtualization enabled running many disparate workloads on a shared server; multicore enabled an uptick in performance and power efficency improvements; Google proved that large scale data centers could be built from commodity technology and operated with amazing economics. The rapid evolution on mobile access and devices are transforming the requirements for enterprise applications providing a high-value motivation to change application front end architecture, just at the right time to leverage Cloud implementations.

There is hype and fluff in the Cloud discussion but underneath that there is a remarkable confluence of independent technology waves. In this case the substance is far greater than the hype.

Intel's AppUp: A Hybrid Cloud for SMB (a clever example of the power of virtualization!)

Like EMC’s VCube introduction last month (VCubes — a really clever idea) Intel’s introduction of AppUp -- an SMB offering leveraging virtualization -- is at once pretty obvious and quite profound. To understand this duality you first need to have a realistic picture of the discouraging nature of SMB IT today. For example, in an auto dealership you’ll probably find many different Windows applications each running on a separate PC or small server. If one of these servers fails or needs an upgrade it means ordering and configuring a new and matching server, carefully reinstalling the application and then grabbing the old data. The power of the server is rarely an issue. With today’s multi-core systems, a single small server typically has more than enough power to roll up and run all these applications on a single, shared machine. The question has always been how to into a more attractive and modern Cloud-like vision with as little disruption as possible.

Intel's AppUp is new and innovative model for delivery IT services to an SMB, developed and operated by Intel, and delivered to SMB customers by Intel business partners (including SP’s, MSP’s and ISV’s). Technically, the customer gets a small, virtualized server and runs the applications they choose (selected from an Intel “App Store”) as virtual machines on that server.  Software and updates are delivered over the network by Intel.  The user’s data resides locally on the server (providing robustness in the case of communications failures) and is backed up into the Cloud.  The overall system is maintained and supported by the business partner.

The AppUp business model is unquestionably modern. There are no up-front costs for the customer (the server is owned by the business partner). The customer pays for the apps on an ongoing, as used basis, based on the number of enabled users.

So on the one hand, AppUp is a straightforward usage of existing ideas, but on the other hand, it’s revolutionary.

• From the point of view of the customer, application selection and installation is dramatically simplified: click and go; and hardware maintenance is greatly simplified as well since the applications share a server, and since any dependence of applications on specific server hardware is eliminated.
• From the point of view of the SP, MSP or ISV who provides support to the customer, the task is greatly simplified. The hardware is generic, based on an Intel definition, and then implemented and sold by multiple hardware OEM’s.
• Usage is moved to an on-demand, pay-for-what you use model. All costs are fees for use rather than up-front capital outlays. There is no need to buy hardware or software up-front. In Intel’s business model, the business partner owns the hardware and lends it to the customer.
• From the point of view of the ISV it’s a way of leveraging many of the benefits associated with the Cloud (no setup, on-demand access, pay for what you use) without changing the software architecture.
• Since the applications execute on the customer’s premises, it avoids entirely the issues of having sensitive data in the Cloud as well as making the solution robust with respect to communications failures.

AppUp is another good example of how a relatively simple use of virtualization enables quite dramatic changes in IT.

VCubes — a really clever idea

One of the announcements at EMCWorld was the packaging of the Documentum software offering into a virtual application. On the one hand, this seems very unsurprising given that VMware is part of the family. But on the other hand, this transforms "enterprise" software dramatically. Historically selling enterprise software (a complicated application) was just nasty. Every customer had a different set of infrastructure elements (O/S choice, storage choice, networking design, ...) and the application had to be carefully adapted to all of that, after which you had a custom application and another supported configuration for the vendor to deal with. Documentum certainly had those issues. Packaging it as a virtual application means creating a composite application, including the underpinning infrastructure elements that then run on a VMware fabric. If the customer has a VMware data center they can just plop it down and run it (of course it's really more complicated but still much less complicated than what it replaces). Or the application can be run by a service provider that provides VMware services. EMC also mentioned the possibility of selling the application along with a hardware system to run it (a stand-alone VBlock perhaps). The hidden gotcha in the idea is all of a sudden there is a new set of software infrastructure (operating system, middleware, ...) that must be kept up-to-date with patches, especially if vulnerabilities are discovered. If the VCube platform was to be really simple to deploy, that problem had to be solved too (and was). The VCube includes a "maintenance socket" (my term, not theirs) that can be connected back to a maintenance service provide by EMC. EMC needs to know all the patches and updates for the components of the VCube and how to correctly sequence them for minimal disruption. Interestingly this is just what VCE is learning to do at lower levels of the stack. This definitely feels like an important element of the future.

The Exchange in the Clouds

(If you squint your eyes just right it looks like a bull!) The NYSE, in conjunction with VMware and EMC, announced a Financial Services Cloud offering. This is a really good example of what IBM introduced generically some time ago — an industry specific cloud. What's significant about this announcement is the participation of an industry segment leader (NYSE) and that it's built on VMware/VCloud, not what IBM would have used, not built on Microsoft Azure (although in some ways it is probably the more suitable platform). As I noted in the last issue with respect to Hyper-V, VMware is the putative winner in data center virtualization. Given that the enterprise data center is getting more enterprise attention than the public cloud, it stands to reason that VMware is going to have significant influence in what gets build in public clouds (perhaps multi-tenant clouds would be a better term), and that is certainly the case today. Many other vendors want to ignore this because it may not speak well to their future success (e.g. as a security provider) but that won't make VMware's success and influence go away.

Service-Oriented Is a Big Deal

VMware's vCloud Director (née Project Redwood) represents a really important step in utility computing because it enables the customer (the SP's that VMware introduced with, but more importantly enterprises converting their data centers to "private" clouds) to create service-oriented solutions "out of the box" rather than having to piece them together with a set of different products. There is no doubt in my mind that service-oriented IT — where the customer drives the system with minimal human intervention — is an important endpoint for most enterprises. To me it's a question of which enterprises get to SoIT in what timeframe, not whether it's going to happen. Ignoring that important timing question for a moment, consider what happens in a service-driven system (where the fast majority of "IT" requests are satisfied by automation) the business can deploy a new instance of an application by pushing a button assuming that enough aggregate resources exist and the business unit has budget. The result of all of this is that IT becomes a lot like Safeway — there is no way either can predict what a customer will ask for next and it's important that essentially all reasonable requests are met, while at the same time maintaining efficient and cost-effective operations. Accomplishing this demands that the infrastructure in turn be managed by automation — not by humans. The "service" definition needs to include all the salient details required to adjust the entire infrastructure to accept the new load. People don't disappear but rather than being "at the controls" they become the overseers, responsible for finding out where assumptions don't work and making things better over time. I've believed for many years that finding a graceful way of getting the humans out of the minute-to-minute operation of the (say) network would be an important and necessary step if the network quality of service was to continue to improve. Service-directed platforms now become a driving force to make these important changes actually happen.

vCloud versus Azure

One of the enduring back stories in virtualization is the role of Microsoft. Every time we ask VMware customers if they have counted out Microsoft the answer is "not yet" because these customers all have strategic relationships with Microsoft and most of the applications and services virtualized to date are based on Microsoft technology. Every quarter Microsoft reports gaining a larger hypervisor share, and yet from a VMware perspective, aren't vCloud Director and vSphere yet one more demonstration of how far VMware is ahead? Can we finally call the question from these initial returns and declare VMware the likely victor? Not really, because VMware and Microsoft are really running for different offices. As noted earlier in my TechEd notes, Microsoft's overarching goal is to convert their software offerings into continuously evolving services, delivered optionally locally or from the Cloud (to move beyond the "release" and "upgrade" model today). Microsoft has a more detailed understanding of the complete lifecycle costs of their customers than does VMware which should not be surprising given Microsoft's product breadth and their long standing study of infrastructure costs and benefits. And from that perspective, believes that eliminating the customer costs for the upgrade and maintenance of software "products" is the key driver they need to respond to, assuming that customers are basically driven by cost minimization (which of course is the primary value proposition of virtualization and the Cloud). VMware has clear leadership building utility computing but ones that are built on conventional software release and install models (it's vSphere 4.1 now). When you get an Azure Appliance you get your own patch of continuously evolving and remotely maintained Cloud, presumably with very different customer economics if you add up all the right bits. Which is the right strategy, assuming that each company can succeed in fleshing out its vision? Ultimately customers will vote with their budgets - it's all about money, presuming the products can be built. Time will tell.

VDI 2010

This part of the VMware strategy seems to have changed a lot in the last 12 months although the VMware spokespeople didn't like it when I said that. One of Paul Maritz's earliest acts was to amp up the VDI investment a lot, believing it would seem that desktop management was ripe for the same kind of revolution that had driven the first wave of data center consolidation, and that VDI could be an important kicker to VMW revenues. I've always been personally skeptical, believing there to be good markets for server-based PC virtualization, but relatively small ones (we even have a two year old report to that effect J ). Citrix's view has been different, benefiting from their longstanding leadership in a broad "application delivery" market, and from that context viewing virtualization as an important new tool to work with but not the second coming. In Mark Templeton's Synergy 2010 keynote he painted a vision based on getting to your "stuff" (my term, not Mark's) from any device anywhere in the world in a fashion that optimized performance while preserving security. The Maritz view is very much in-line now, emphasizing user device support more than the wonders of server virtualization. To be fair to VMware, no one reasonably anticipated the impact of the iPhone and iPad on corporate IT. Templeton and Citrix have promoted the concept of "consumerism" for quite a long while, but Apple moved that from a strategic alternative (something a company might choose to do) to a tactical imperative (in the last quarterly call, Mark made special mention of how the iPad is starting to really disrupt corporate IT). VMware seems to agree that application delivery (your stuff from anywhere on any device) is the right focus (not their words). If that's the mission then Citrix's strengths in the battle are more self-evident now. It's going to be a good (and productive) war that the consumer will ultimately benefit from!