Last fall, I suggested that Hewlett-Packard's version of Unix--HP-UX--could be Intel's ticket into the data center. Why HP-UX and not Windows? Also last fall, Microsoft CEO Steve Ballmer was candidly noting that Windows had thus far failed to penetrate the data center and that, going forward, this penetration would be his biggest challenge. But now that challenge is looking more like a window of opportunity as long as Microsoft addresses the biggest and most outstanding concerns of data center managers.
People and performance are the two concerns that ZDNet's readers have voiced most often about Windows' readiness for hardened data center applications. On the people side, a lot of IT executives have written to me about how Microsoft-certified engineers are a dime a dozen; finding one that actually knows how to deploy Windows, let alone any operating system, in a data center environment, they say, is like finding a needle in a haystack. On the performance side, a big concern is Windows' ability to survive the rigors and demands of the data center in the way other industrial strength systems can.
Microsoft is aggressively addressing both issues--and not a moment too soon. Given the high-level of interest in the Web services architecture and the fact that mission-critical Web services will have to exhibit the reliability and performance of other data center applications, if Windows has a real shot at the data center, 2003/2004 will be the period it gets to stake that claim. If it doesn't, the window of opportunity may close forever.
Why Microsoft's fortunes could change
One reader who wrote to me about my series on mainframe Linux (one of the very real alternatives to Windows in the data center) pointed out that no operating system is magically equipped to survive the rigors of the data center the day you plug it in. Making that happen, regardless of OS, takes the right people-- people who know how to build successful reliability/failover plans and do capacity planning.
Microsoft is working to assure better availability of capable talent that's equipped to deploy in data centers not only Windows, but Web services powered by Windows. One of the distinct advantages of the Web services architecture is that Web services clients are totally insulated from the operating system that drives the service. This means the outward-facing software interface is OS-independent and that, with the right talent, a Windows-based solution could easily be substituted for a Unix- or mainframe-based solution without causing so much as a hiccup to the clients.
One choice that data center managers already have is to approach a consultancy that knows both Windows and data centers. Microsoft and consulting firm Accenture have contributed the pedigree of both their companies to a joint venture called Avanade. Michael Condon, a partner at Accenture, and Avanade's e-business director Brian Harp explained to me why they thought Avanade is the "go-to consultancy" for companies considering the deployment of Web services using Windows .Net. According to Harp, Avanade has the crème-de-la-crème of Windows engineers and they subscribe to Accenture's official deployment methodology that Condon says has produced consistent results for all of Accenture's clients. If what they say is true, than it could be the sort of marriage that results in some sorely needed design wins for Windows in the data center.
A data center version of Windows 2000 Server already exists. But to drive Web services, the .Net framework must be added to it. Microsoft recently went on tour to remind journalists that Windows is indeed an application server and should be considered as such in situations where new systems demand an application server. The .Net framework consists of the Common Language Runtime (CLR) as well as several other technologies like ADO.NET and ASP.NET that bridge the CLR to the distributed object, Web serving, message queuing, and directory infrastructures that are already a part of Windows. But with Web services still in an embryonic state, Microsoft has some time to fortify its application server offering in a way that doesn't make .Net seem like an afterthought to Windows.
Somewhere near the end of this year, Microsoft expects to ship Windows .Net Server. This successor to Windows 2000 server will hold two big attractions for data center managers. First, .Net will be more deeply woven into the fabric of Windows. Second, the server version of Windows has continually matured over the last decade. Not only has each successive version taken on more enterprise-like features to deal with issues such as security, fault-tolerance, and scalability, but also Microsoft has learned a great deal from supporting enterprise customers that have already been pushing Windows to its very limits.
Finally, all of this comes at a time when Moore's Law continues drive the performance of low-cost Intel hardware into a range where data center managers will simply have to consider the option. But with Linux coming on strong (and with Intel and IBM backing Linux as much as, if not more than, Windows), the big question will be "what OS?"
What do you think? By the time Web services start to get some traction, will Windows have its opportunity to shine in the data center? Let me know. TalkBack to me, or write to me at david.berlind@cnet.com.
