On January 31st 2013 Intel issued a statement on their website, which basically nullifies their bold plans for their Itanium line of processors, which were only publicized a little over three months ago. Accompanying the launch of the Itanium 9500 series codenamed Poulson, Intel laid out a roadmap where the follow-up dubbed ?Kittson? would be shrunken to 22nm process technology and share a common platform (i.e. socket, chipset, etc.) with the x86 high-end Xeon E7 line. Those plans are now null and void.
The statement reads:
?Intel has updated the definition of the next generation Itanium® processor, code name ?Kittson?. Kittson will be manufactured on Intel?s 32-nm process technology and will be socket compatible with the existing Intel Itanium 9300/9500 platforms, providing customers with performance improvements, investment protection, and a seamless upgrade path for existing systems. The modular development model, which converges on a common Intel® Xeon®/Intel Itanium socket and motherboard, will be evaluated for future implementation opportunities.?
So, instead of a common platform, the next Itanium product will be rather a very minor upgrade to the current lines with backwards compatibility with the existing infrastructure. Given the stagnating development of Itanium a common platform with the much higher volume Xeon E7 line would have made a lot of sense to continue support for the IA-64 architecture CPUs while keeping costs in check and not falling too much behind in terms of platform technology.
An snap shot of an Itanium 9500 Wafer
While the statement leaves Intel a back door to continue with the roadmap as laid out at some point in the future, it’s quite safe to say that this is the final nail in the coffin of Itanium. By the time Kittson ships ? if ever ? 32nm process technology can be considered outdated. By the time Intel will start 14nm manufacturing at least in the consumer space and shortly thereafter in the mainstream server segment. The big cuts on the platform side basically means it is not moving forward either but rather frozen at the state of 2010, when Intel introduced the Itanium 9300 ?Tukwila? line. Currently we can only speculate whether Intel will incorporate CPU architecture update in the revised Kittson, but an educated guess might be that it will only include minor errata fixes at best.
It took quite some time for IT news sites to pick up this story. While Intel made the original announcement almost three weeks ago, only now a lot of stories are popping up about it. We admit that we didn’t cover it any earlier either, but maybe that is also a testament to how unimportant Intel?s Itanium line got over the course of the last years.
Our last prominent coverage of the processor architecture was back in 2011 when HP and Oracle had a public dispute over software support on the Itanium platform. The dispute eventually got settled in court in August 2012, where it was ruled that Oracle has to provide software for Itanium-based servers for HP as long as HP decides to make them. Oracle appealed the decision in October last year but their petition was denied on January 31st 2013 by the court. As a result Oracle already resumed to offer Itanium versions of their software. Still they continue to uphold their original statements on their website along with the updates on the situation.
Regarding Intel’s announcement, HP issued the following statement to the press:
?On January 31, 2013, Intel posted a statement on intel.com, providing an update around its plans for "Kittson", the codename for Intel’s next generation Itanium processor. HP and Intel have a long history of working together on Itanium processors and Integrity systems to deliver mission-critical infrastructure for the world’s most demanding applications.
HP is committed to the Integrity product line, including ongoing innovation and development of a new line of Kittson-based Integrity systems in the future, working closely with Intel. The recent statement by Intel has no impact on those plans or on HP’s ongoing commitment to our mission-critical customers.?
This is what can be called a typical example of corporate PR damage control. Those two paragraphs are meant to reassure those customers who haven’t yet jumped ship that everything is continuing as planned. However, given that both HP and Intel didn’t even bother to do proper press releases but released some sneaky statements buried somewhere on the internet only, the evidence is piling higher and higher. Itanium used to be on life support for the last years and now the plug has been pulled.
It is an open industry secret that HP subsidized Itanium development at Intel for some years. One possible explanation is that they cut those subsidies which results in Intel taking the logical step of axing further development. It is unclear whether this is related to the shenanigans with Oracle. We do know however, that HP is working on an exit strategy for their Unix business which is supposed to be moved to Intel’s Xeon line of processors in the long term, as HP CEO Meg Whitman alluded in an interview with All Things D.
And the rest as they say is history. Back in 2001 Intel had a bold plan when they introduced the Itanium line, which was poised to gradually replace the x86 architecture, which back then was incapable of a 64-bit mode and thus not able to address more than 4GB of memory which increasingly became a crippling limitation. With the introduction of their Opteron line, AMD extended x86 with 64-bit and thus enabled it to cater to the server segment again. Intel eventually licensed the instruction set extension back from AMD which effectively denied Itanium entry into more mainstream market segments. It was moved into a very specific niche of high-end Unix mainframes with HP as the sole effective customer. Now support for existing systems will be going on for years to come, but the journey has come to an end.