Intel's Tukwila slips yet again

Intel has slipped out a revised schedule for its next-generation Itanium processor, code-named Tukwila. Again. This time it's into 2010.

Intel released a statement Thursday on the schedule changes. It reads in part:

During final system-level testing, we identified an opportunity to further enhance application scalability best optimized for high-end systems. This will result in a change to the Tukwila shipping schedule to Q1 2010.

In addition to better meeting the needs of our current Itanium customers, we believe this change will allow Tukwila systems a greater opportunity to gain share versus proprietary RISC solutions including Sparc and IBM Power. Tukwila is tracking to 2x performance vs its predecessor chip. This change is about delivering even further application scalability for mission critical workloads.

That may be true. However, the fact remains that this is yet another delay to the program. This will put Tukwila's introduction more than two years after the debut of the current "Montvale" generation--which itself was a delayed and modest speedbump to "Montecito"--and one that Intel barely announced publicly.

Tukwila has had an especially bumpy history. This generation of Itanium processor began life as a chip project code-named Tanglewood and was said to be envisioned as a radical multicore design by the ex-Digital Equipment Alpha engineers who worked on it.

First, Intel changed the code-name to Tukwila after the Tanglewood Music Festival complained. This was back in 2003--to give you an idea of how long this particular project has been weaving its way through development. At that time, it was slated for something in the neighborhood of a 2007 release.

Then the chip apparently went through a variety of significant design changes. It will still be the first Itanium to sport Intel's serial processor communications link (QuickPath Interconnect--QPI) and integrated memory controllers. Those are both major enhancements, but otherwise Tukwila is a more conventional quad-core evolution of current Itanium designs. It will also be manufactured with a 65-nanometer process instead of the denser 45-nanometer process already used by the newest Intel Xeon CPUs. Along the way, the chip's schedule has been publicly pushed back a number of times, now to early 2010.

As a practical matter, delays to Itanium matter less to Intel and the server makers that use it (meaning Hewlett-Packard first and foremost) than in the case of x86 Xeon, where a delay of a few months can have a major revenue impact--vis-a-vis Advanced Micro Device's Barcelona.

Buyers of high-end servers like HP's Superdome and NonStop value vendor relationships, reliability, and a wide range of enterprise-class capabilities far more than they do the last drop of performance. HP has done a good job of things like leveraging its c-Class BladeSystem infrastructure for its Itanium-based Integrity servers and putting together systematic go-to-market programs with partners such as SAP.

Nonetheless, at some point, ongoing delays have to hurt competitiveness--especially given how IBM's Power systems have been hitting on all cylinders the past few years.

[NoReligion]

How did they come up with the name 'Tukwila'. I'm sorry, especially to any of my fellow Puget Sounder friends who might be from Tukwila, WA. But that name in the Puget Sound region is evocative of ...less than perfection, and more funny, of "the nooner" ....Just ask someone from Seattle.

[Vegaman_Dan]

Tukwila... the name of the southern junction of I-5 and I-405 in the Seattle / Tacoma area. An area that is known pretty much only for the Southcenter shopping mall and horrendous traffic jams. It is perhaps one of the most congested and slowest performing areas of the entire state.

And Intel wants to name a chip after this?

Yeah, I don't get it either.

[ghaff]

Lots of Intel chip codenames are based on West Coast river, city, and other place names. Don't know why Tukwila in this specific case but I'd point out that Intel has used up a lot of codenames over the years so a lot of the scenic places may have been used by now.

[gerrrg]

The naming convention is derived from the practice of using local names from the city whence the engineering design comes from, although I think they maybe should have named it Bellevue, instead.

[Tod Smith]

I think the Tukwila could take off if MS makes an X86 EMU for it.

[medezark]

Is the Itanium line of processors still relevant?

[ghaff]

Yes. It underpins HP's Integrity line--NonStop and Superdome etc. (the latter of which run primarily HP-UX but also OpenVMS and to some degree Linux and Windows). It's actually a growing set of products--competitors to IBM's POWER.

[Alphaman63]

Yes, very relevant. If you can show me a single x86 system capable of having 256 TB of RAM and 273 GB/sec memory throughput, then it might not be. Xeon and Nehalem are about as relevant as an ARM processor when compared to current gen Itanium systems.

[Mr. Dee]

Seriously, Intel should just sink that ship and put all passengers on a life boat to XEON. This processor platform was planned as the successor to x86, reading into its history, it was just a bad decision from the start. Even the first designs, which started back in the late 80's didn't come to market until about 2001 I believe. The fact that both HP and Intel, who have invested so much research effort into it are not seeing any significant benefits or wide industry adoption should just realize that RISC belongs to IBM Power and SUN's Sparc. With Oracle 'the devil Larry' committing to Sparc and Solaris, its gonna be a tough fight ahead for Intel and HP.

[ghaff]

It's a matter of software. NonStop, HP-UX, and OpenVMS are all tied to Itanium. You can't just up and move the operating systems and thousands of apps to Xeon. And binary translators and whatnnot aren't much use for enterprise-class apps which these are.

[Alphaman63]

Sorry, Mr. Dee, but you've really shown your ignorance of server processors. Your statements sound like you learned everything you need to know about Itanium from WikiPedia. Intel's Xeon processors aren't even in the same class as Itaniums. Itaniums, while originally targeted as a possible replacement for the CISC family of Pentium processors, has ALWAYS been about EPIC being a competitor to RISC.

Intel and HP have plenty of customers on Itanium now (myself included, being brought kicking and screaming into the fold) who scoff at the thought of going to Xeon. And Tukwila, with its Alpha-based design, will be worth waiting a few more months for. It may not be pertinent for x86 fans, but it certainly is for those of us running big iron where downtime is NOT an option (Haff is spot-on in his statement regarding performance vs. reliability in this article). Delays like this are de rigueur in this market, because of the specialized nature of the beast.

I think there's more to this story -- engineers don't get to decide to tweak performance and slide schedules during final system level testing. Marketing would flat-out reject that, and plan the performance boost for a down-the-road upgrade. No, I'm much more inclined to think they found something very broken on the die, and wound up having to do a significant repair job to the mask. They may be using this as an opportunity to add a performance enhancement, or the problem may have resulted in a significant performance degradation, but either way, I think they're fixing something broken to get this "improved performance".

But I wouldn't have wanted to even been a fly on the wall in the meeting where engineering had to tell marketing they were slipping!