By Charles King, Pund-IT, Inc. – I’m delighted to welcome Charles as a contributor. This piece was published in the PUND-IT newsletter.
In a blog, Dan Reger, senior technical product manager for Microsoft’s Windows Server group, announced that Windows Server 2008 R2, SQL Server 2008 R2 and Visual Studio 2010 will be the last Microsoft products to support Intel’s Itanium microprocessor architecture. Mainstream support for Windows Server 2008 for Itanium-based Systems (and R2) will end, in accordance with Microsoft’s Support Lifecycle Policy, on July 9, 2013, while extended support will continue until July 10, 2018.
Reger’s post incited a smattering of articles and Tweets over the weekend, including speculation on the health of Microsoft and Intel’s alliance and Itanium’s long term prospects. So why is Microsoft ending support? According to Reger, the decision rests in the “natural evolution” of the 64-bit x86 (x64) architecture to support mission-critical applications and workloads. In addition, he pointed to the blazing performance of Intel’s new Xeon 7500 (Nehalem EX) and AMD’s Opteron 6000 (Magny Cours), along with a new NEC TPC-E server benchmark record, as proof points for the decision. In other words, Microsoft believes that mission critical computing is heading inexorably toward x64 and decided to act accordingly.
Good enough but does this make sense? Yes and no. On the plus side, we agree that the future of x64 has never looked better. The new Intel Xeon and AMD Opteron processors are quite simply the best and most able x64 server offerings either company has ever produced. More importantly, Intel’s and AMD’s respective multi-core technologies look extensible for a long time to come. In other words, the price/performance engine which has driven the remarkable success of the volume server market and increasing demand for high-performance x64 systems has a full tank of gas and a long, straight road ahead.
But at the same time, Reger’s suggestion that the future of mission-critical computing has an inevitably x64 cast seems both a blithe overstatement and wishful thinking on a grand scale. While demand for x64 continues to grow, the market for systems based on Itanium and RISC platforms, including IBM’s POWER and Oracle’s SPARC, was about $12 billion in 2009. Not chicken feed, by any measure, and the figure does not take into account the billions in annual revenues generated by IBM’s and other vendors’ mission-critical mainframe systems.
So by dropping Itanium support is Microsoft stepping away from that market? Not at all. Just approaching it in what the company deems a more cost-effective manner. The fact is that Microsoft’s Itanium efforts (beginning in the mid-1990’s) have never really resonated commercially. According to recent survey data, less than 5% of Itanium systems sold in 2009 supported Windows Server, while UNIX- (particularly HP’s HP-UX) based Itanium sales continued to grow. Microsoft’s move is a significant reality check for boosters who continue to posit Itanium as the foundation of Intel-based mission critical enterprise computing. If any further evidence were needed, Microsoft’s decision cements x64’s role at the center of those scenarios.
So where does Microsoft’s announcement leave Itanium and the server vendors who continue to support it? Mostly unfazed. Regardless of whose market data you consider, HP owns 90%+ of the Itanium market and leverages the platform across a variety of common (HP-UX, Linux, Windows) and legacy (Open VMS, NonStop) operating environments. The size of the company’s relative commitments to Itanium makes any radical changes in HP strategy doubtful. In addition, given the company’s close relationship with both Microsoft and Intel, one expects HP will easily step up to support its partners’ shifting strategies.
The same is even more likely for smaller players, like Bull and SGI, which leverage Itanium in Linux-based systems. In fact, they and the vendors who do offer Windows on Itanium systems, including NEC and Fujitsu, are also enthusiastic supporters x64-based technologies. One expects that by 2018, many if not most of those vendors’ Windows on Itanium customers will have long since migrated to x64.
No matter how dramatic Microsoft’s move might have seemed, its exit is not likely to significantly impact the company’s revenues or overall Itanium platform sales. Most importantly, offering extended support for Itanium until 2018 means that the company’s customers will not be left in the lurch. Overall, Microsoft’s decision qualifies as an orderly exit from a market that simply never lived up to expectations. Bottom line: the biggest news here is that there is no real news.
© 2010 Pund-IT, Inc. All rights reserved.
About Pund-IT, Inc.
Pund-IT emphasizes understanding technology and product evolution and interpreting the effects these changes will have on business customers and the greater IT marketplace.
As it happens, I disagree with Charles that “the biggest news here is that there is no real news.” It’s an admission of failure on Microsoft’s part on the one hand to crack the data center database server market with Windows Server and SQL Server on Itanium, based on the penetration numbers Charles provides.
Microsoft’s benchmark strategy seems to confirm this. It continues to be alone for the past year publishing TPC-E benchmarks; its first Windows Server R2 one was published in November, running the first SQL Server R2 benchmark. Other SQL Server R2 benchmarks followed; none have been Itanium-based. Interestingly enough, two Microsoft TPC-Cs have been published on Windows Server R2 this year; again, neither was on Itanium, and both used SQL Server 2005, not the coming release. The AMD Opteron-based one, published April 8, 2010, comes in at #4 on the price/performance chart here. It seems evident that the engineering money Microsoft spends for benchmark optimization is elsewhere. And so, from the top results of the last few years, is everyone else’s – at least as far as database performance is concerned.
It’s clumsy from a communications perspective that Microsoft announced the end of support date for Itanium just when it is about to do a major launch with SQL Server R2. Charles and I agreed in a recent discussion that the timing, just three months after the joint announcement the two vendors made about their Frontline Partnership, reinforces our skepticism about that announcement as well.
– Merv Adrian