It turned out to be a very Cloud-themed, meeting-oriented 3 weeks for me. First of all, Cisco, EMC and VMware (with a little help from Intel) announced a set of cooperative efforts aimed at accelerating the uptake of large-scale, virtualized computing. Then EMC had an Industry Analyst Meeting that was very much focused on the "Private Cloud." Finally, at Microsoft's PDC in Los Angeles, Microsoft's Cloud services (Azure) was at the center of much of the discussion. I'll talk about each of these briefly in this issue. They each see the Cloud through their own unique lens. Each is quite interesting (IMHO).
Cisco hosted the first shoe dropping, and John Chambers presided. Cisco and EMC have been at work for some time developing "blueprints" for chunks of virtual infrastructure (servers, network and storage) called VBlock's. They discussed three configurations (VBlock0, 1 and 2) ranging from substantial to quite large (minicomputer to mainframe price, in legacy terms). Each VBlock comprises a configuration of Cisco's UCS integrated server and network offering, mated with a suitable EMC storage offering, using VMware VSphere virtualization software. Cisco uses the term "blueprint" normally to mean a configuration that is very precisely specified (physical and logical configuration) and thoroughly tested. If you want to buy a piece of virtual infrastructure of some capacity (measured in virtual servers) this is a way of getting one that works and doesn't require much customer learning to install and commission correctly. Cisco and EMC motivate this effort as follows: Both companies are convinced that moving data center operation to a highly virtualized structures results in very significant savings in data center operation, both CapEx and OpEx (not to mention higher quality operation). There is enough evidence of the simple benefits of server consolidation to make this quite plausible. When this vision is presented to prospective customers, the customer typically asks "Tell me what I have to do." VCE (the consortium) and VBlocks are part of the answer — here's a template for a known good solution. But the effort doesn't stop there. There is also a Joint venture called Arcadia. Arcadia will take the next step and deploy a VBlock for a customer, and operate it for a bounded period of time. Our "knowledgeable, unnamed sources" say that up until the announcement, the structure of Arcadia was under debate, and creating a real product company was considered (moving development and manufacturing resources to Arcadia), but in the end this unique form of JV was chosen. If the arrangement can be made to work, it will be an elegant solution. Underneath the covers, Cisco, EMC and VMware are working to integrate their service management systems, so a VBlock solution can be serviced by an EMC, Cisco or Arcadia customer with only "one (vendor) throat to choke." Of course the world is complicated by things like discounts, sales commissions and account management, so we'll just have to wait and see how it works out in practice. Cisco's motivation seems reasonably straightforward. The Nuova/UCS solution does an elegant job of integrating the networking and server components but left storage selection, integration and operation to the customer, and all these pose technical and cultural challanges. UCS acceptance and deployment will be made easier and accelerated by the availability of VBlocks and Arcadia. We also heard that Cisco thinks that these product blocks will be attractive to the integrators who compete with the in-house groups of HP and IBM, so go-to-market values also play into their thinking. EMC was very clear that they will be a technology company: they have nearly 2,000 consultants in their professional services group but that a much smaller part of the business than with HP or IBM, supporting the notion of gaining market strength though the unaligned integrators.
I missed last year's EMC Industry Analyst meeting, and this one was completely different from 2007. Two years ago it felt like EMC was a manufacturer of complex storage boxes that dabbled in software and applications as needed. In 2009, from Joe Tucci down, EMC seems very much like a systems company, and one that understands (at least strategically) that software is very much a key part of its future. To be sure, EMC is still very much a storage company and makes money by selling boxes. But a modern storage system is mostly complex software, and (it turns out) sometime in the past EMC moved to using X86 instead of alternatives like MIPS or PowerPC. In a virtualized world, it may well make more sense to package storage functionality as virtual machines than as boxes in the infrastructure. In fact, Chad Sakac, EMC's VP of VMware Technology Alliances, in his demo showed off some virtual Clarion systems running on the virtual storage fabric which is much easier (still not simple) if it is already running on an X86 platform. In his presentation, Brian Gallagher, the head of the Symmetrix business, let drop that he had 1,600 engineers, all software engineers (and all writing X86 code to boot). EMC seems to clearly understand that they need to make their customers succeed with storage for virtualized environments, even if that journey includes selling more software than hardware and aggressively bringing the cost of storage down. In the last year EMC introduced FAST (fully automated storage tiering) that lets flash and SATA drives be mingled with the traditional expensive drives in a Symmetrix system. The results can be dramatic: cheaper, faster and lower power. Next EMC has their sights set on what they call storage "federation" — how many block storage systems can be linked within data centers and across data centers including capabilities like storage motion to match application motion, and FAST and storage federation are essentially just software. Managing data within and across public and private clouds is certainly a key part of the vision of utility and cloud computing. In contrast to Cisco with significant businesses in both enterprise and service providers, EMC's bread is very much buttered by Enterprise sales, and their current focus is very much on the journey to the "private" cloud — the use of virtualization to evolve enterprise data centers to utility, self-service computing. I personally agree that this is the most exciting story (not the public cloud). EMC seems to have set the stage for a remarkable metamorphosis from the big box past to something quite different. Of course setting the stage and giving the performance are quite different, so only time will tell.
Having been infused with discussion of the private cloud, it was time for a regular pilgrimage to MS PDC in LA (a road trip down I5!). Microsoft's offering (and story) is much more complex: they are a huge software company with a wide spectrum of offerings; they have been in the online business for years; they have a long term investment (and large business) in systems management. Ray Ozzie's first big keynote was at the MIX conference in 2008, then Azure was discussed at PDC last year, virtualization and operational management is normally the topic of MMS, and of course TechEd is where the enterprise products are discussed. Utility computing and clouds span all of these venues. In the private cloud discussions, Microsoft's contributions and capabilities are usually deprecated. VMware is clearly the leader in enterprise virtualization and after all, cloud computing is high-end, complex virtualization, right, and therefore clearly not Microsoft's strength. But of course, it isn't so simple. VMware's big recent strategic move was the acquisition of Spring Source and the Spring framework, and Microsoft was deeply into frameworks (first MFC and then the .Net framework) long, long ago, and Microsoft has a long term investment in system management as well, with a systems management business comparable to all of VMware. And then of course, Microsoft has been building very large scale Web properties including Azure. So who exactly is to say that Microsoft can't do private clouds and utility computing? Certainly not Microsoft. It wasn't the focus at PDC. Azure, Windows 7 and Silverlight were. Azure goes live as a business in January 2010 and it looks really interesting based on what's been done so far. Specifically, in a parallel effort, Microsoft has been developing SQL Services for the cloud, now rebranded as SQL Azure, and at PDC they announced "Dallas" — a project to build relationships with all sorts of interesting data providers, and make these data sets available, on demand, priced by the sip, on Azure. And then the Live Labs team talked about Pivot, a tool that allows for the visual browsing of large data sets using the Deep Zoom technology (also integrated with Silverlight although Pivot isn't yet a Silverlight application). It's sort of like the BestBuy eCommerce interface on your favorite data set (you have to see a demo to really get it I'm afraid). Quest Software demoed an Active Directory backup service built on Azure (it really highlights the value of all the Windows infrastructure compatibility). And then of course all of the Office/Exchange/SharePoint stuff is increasingly wound into this model as well. And then there is Silverlight, which grows more powerful by the day (literally) — some 26 months since it was introduced they are beta releasing Silverlight 4. The Silverlight team thinks that usage will start to explode when they can reach 50% PC penetration (Flash is still higher). At the Silverlight 3 launch this summer they were at 33%; at PDC they said 45%, so 50% by the end of the year looks quite possible (Microsoft gains installation share the good old fashioned way by doing things like NFL Sunday Night Football and the upcoming Victoria's Secret fashion show). The point is that if you add up everything Microsoft is bringing to bear its pretty amazing and seems to sort of fit together in a tantalizing way. The back story is that Microsoft is working to integrate the Azure Fabric Controller (what was developed beyond Windows to make a Cloud version of Windows) with System Center as the basis of a private cloud technology. It was hinted that they are in Beta with significant customers (since Azure goes business live in a month, the code base is reasonably shaken down). We think all this will come to light in the first quarter or so at Mix or MMS.
Underneath all the fun stuff at PDC, you just can't ignore the importance of Windows 7. GM doesn't get a chance to sell Volt electric cars if they can't succeed selling Chevy's, and the same is true for Microsoft, so Windows 7 was very much in the discussion. In one of the PDC keynotes, Steven Sinofsky, President of the Windows and Windows Live Division, gave some interesting overview comments about building Windows. Two that stuck in my mind: first, the challenge of taking a new college grad and helping them make the transition from writing code for homework assignments to "checking in code for 100 million users." Creating a release of Windows is a unique and daunting engineering challenge, and there is a very interesting side story about how Sinofsky and Jon DeVann fixed the Windows engineering process so the different programs came together better and faster than had been the case with Vista (Sinofsky and DeVann had conspired earlier to perform similar tricks in Office). I'm hoping someday Microsoft will tell this more publicly. Sinofsky's other really interesting comment was about the feedback they got from with Windows 7 beta process, which included the configurations used. It turns out that over 50% of the Beta users had XGA (1024 x 768) displays, while the engineers typically use dual HD displays. Seeing these configurations is a real wakeup call as to what "compatibility" really means, especially if you consider that Windows 7 beta users are hardly technical Luddites. If you're a Windows 7 programmer not only do you have to create code for 100 million users, but you also have to make it work on their computer as well as yours.
