Print 24 comment(s) - last by Calin.. on Feb 28 at 2:55 AM

Current Intel "Montecito" Itanium die
Intel's Itanium and Xeon architectures will have more similarities in the future

Intel’s enterprise Itanium and mainstream Xeon server platforms have remained completely independent from each other, in terms of technology. There has never been platform sharing, chipset sharing or anything else in that regards. However, Intel expects to converge the two server and workstation platforms. 

Pat Gelsinger, Intel’s senior vice president and co-general manager of the Digital Enterprise Group, outlined future convergence plans of Xeon and Itanium product lineups in an interview.

When asked how its Core and Itanium architectures will intertwine, Gelsinger responded “The first realization of that is Tukwila [quad-core Itanium] in late 2008, the next step in the product family, where we move to common system architecture elements, as well as full alignment on design tools and process.”

The upcoming Itanium Tukwila will take advantage of Intel’s common system interface, also known as CSI.

As Intel moves its Itanium and Xeon processors converge towards the common CSI-bus, the processors will share more similarities. Cache architectures of the two processors will become similar in the future, as there are no compelling reasons to have different cache architectures between Itanium and Xeon processors.

CSI-bus aside, Intel’s Tukwila-core is the first multi-core Itanium architecture. Tukwila is also the successor to Montecito, which Intel launched last year.

Intel expects to have common platform development as well. Common platforms will allow OEMs to have a single platform for Xeons and lower-end Itanium processors – easing OEM developments.

Those expecting Intel to produce a hybrid x86-64 and IA64 compatible processor will be disappointed, as Intel does not intend to take Itanium and Xeon convergence that far. “I don't see it getting that far, but I am driving these things to be as common as possible,” said Gelsinger.

Update 2/27/2007: The reference we had to Tigerton supporting CSI in 2007 was incorrect and has been removed.

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

Why is Itanium Still Alive?
By qdemn7 on 2/26/2007 10:22:23 PM , Rating: 2
I have never paid much attention to Itanium, given I really don't care about the server side of things, so I'm ignorant of its appeal. What does Itanium offer for Intel to still keep the platform around?

RE: Why is Itanium Still Alive?
By The Sword 88 on 2/26/2007 10:44:52 PM , Rating: 1
what exactly is itanium?

RE: Why is Itanium Still Alive?
By nurbsenvi on 2/27/2007 1:41:44 AM , Rating: 1
it was massive and sunk in the sea of other products.

RE: Why is Itanium Still Alive?
By CorrND on 2/27/2007 2:20:26 PM , Rating: 3
Itanium was designed around the IA-64 instruction set, a 64-bit replacement instruction set for the outdated x86 instruction set that was originally designed by Intel for the 16-bit 8086 processor in 1978. The thinking was excellent, but the timing and marketing were extremely poor. Wikipedia has a very good synopsis here:

"IA-64 was based on Explicitly Parallel Instruction Computing (EPIC), in which the compiler lines up instructions for parallel execution. Itanium was designed to ensure compatibility with both Intel x86 and HP UNIX applications. During development, it was widely expected that IA-64 would become the dominant processor architecture for servers, workstations, and perhaps desktops, superseding the ubiquitous x86 architecture and providing an industry-standard architecture across an unprecedented range of computing platforms."

RE: Why is Itanium Still Alive?
By sprockkets on 2/26/07, Rating: 0
RE: Why is Itanium Still Alive?
By NT78stonewobble on 2/27/07, Rating: 0
RE: Why is Itanium Still Alive?
By hellokeith on 2/27/2007 11:28:09 AM , Rating: 1
Sprockkets is likely an AMD + Linux (or even worse, FreeBSD) fanboy who thinks opensource software can calm the heart of brutal totalitarian dictators.

RE: Why is Itanium Still Alive?
By sprockkets on 2/27/2007 11:50:28 PM , Rating: 2
Sprockkets is likely an AMD + Linux (or even worse, FreeBSD) fanboy who thinks opensource software can calm the heart of brutal totalitarian dictators.

Yeah, one Prescott based 865G system, one AMD64+ 3000 system, one laptop with a Pentium M chip with a 855GM chipset in it, with all computers running either WinXP Pro legit + SuSE 10.2. Intel hardware is well supported in Linux too so there goes me favoring AMD.

If you want to debate software religiously, then talk to Richard Stalman. Otherwise, some people use linux because it just works better for them. I view microsoft as a screaming whiner because that is how Steve Ballmer acts.

RE: Why is Itanium Still Alive?
By GI2K on 2/27/2007 8:04:35 AM , Rating: 2
Itanium is the living proof that Intel can cut completely with its old line of products, also while Itanium may not sell all the research done for it isn’t wasted, it’s like with the netburst architecture it didn’t serve that well Intel but nevertheless a lot of data was gathered cause of it and now even the Core 2 is probably enjoying of the research done on the netburst.

The same can be said about the 80 “core” CPU they are making now, it doesn’t pay but the research may be used later in other products later on, this is far better than the wait and see state of mind.

RE: Why is Itanium Still Alive?
By ToeCutter on 2/27/2007 9:49:29 AM , Rating: 1
If you want a good analogy of what the Itanium is, just think of the US and the Iraq war; both have spent considerable resources into doing each, and since quitting admits failure and a loss of billions of dollars and would be stupid to stop now, whereas if they keep going more money will be spent with no profit or winning in sight, neither of them know when to quit.


A frighteningly accurate analogy...

RE: Why is Itanium Still Alive?
By Flunk on 2/27/2007 10:06:29 AM , Rating: 4
Actually Itaniums are still the best performing processors that Intel sells. They are used in a lot of high-end servers and super computers. The only problem is that they are not compatible with x86 instructions so Intel has never been able to figure out how to market a cut-down consumer version. Itaniums are in the same sort of class as IBM Power and Sun SPARC processors, enterprise level only.

RE: Why is Itanium Still Alive?
By saratoga on 2/27/2007 11:59:34 AM , Rating: 2
Actually, no. Take a look at Spec scores:

In int, its comparable to a mid range P4 or Core 1. In FP, its about even with or slightly slower then the Core 2 XE. I'd say the real problem with I2 is that the performance was never there. If it had been, people would have made do without x86 support. But if you can get similar performance with x86 (or way better if you need int), why bother?

RE: Why is Itanium Still Alive?
By Calin on 2/28/2007 2:55:35 AM , Rating: 2
The server side of things is Xeon. Itanium, right now, is going for the high performance calculation market (where the programs you run are optimized for every architecture).
Having a non-compatible, very high performance processor with a very slow "compatibility layer" is good if the cost of rewrite/optimisation of programs is small compared to the value of performance you get from that architecture. A 5 million dollar computer, incompatible with every other architecture there is, would be cheaper than a 10 million computer, that can run everything at the same speed, as long as the cost of program porting/library rewriting is in the millions range.
By the way, while some low level libraries are written for the processor in use, others must be optimized based on the best performance to be obtained for the respective architecture - processor type, interconnection type, speed, local memory size for nodes and so on and so on.
The "industry standard" tests with which the big iron builders compete run on architectures (hardware and software) painstakingly optimized for the task in hand.

By Anh Huynh on 2/26/2007 10:03:07 PM , Rating: 1
The shared bus architecture has nothing to do with whether the processor is x86 or IA64. Xeon will remain x86 while Itanium will remain IA64. It will however, allow possible shared chipsets and platforms.

By Zandros on 2/26/2007 10:16:47 PM , Rating: 2
(last I heard, some variants of Tukwila will be well over 2 billion transistors)

Actually, the vast majority of those transistors are the L3 cache. In an old Anandtech article*, we see that the Itanium 2 core have significantly fewer transistors than a NetBurst Xeon. I think it may be 20 million compared to 80 million, but you can look it up for your self.

It is good to see that there still is life in EPIC and the Itanium.

Also of note is that POWER5+'s L3 cache is off die, which makes it seem smaller, when it might not be the case.


By killerroach on 2/26/2007 11:59:06 PM , Rating: 2
Agreed completely, although having the L3 off-die should do wonders to help yields (granted, at the expense of added latency). However, considering the price that a Montecito-core Itanium sells for, I think that's hardly a concern that Intel has at all.

By killerroach on 2/27/2007 12:05:27 AM , Rating: 2
In reference to what was mentioned about transistor count, the breakdown of the 1.72 billion transistors in a Montecito processor, from the Wikipedia article:

core logic — 57M, or 28.5M per core
core caches — 106.5M
24 MB L3 cache — 1550M
bus logic & I/O — 6.7M

So, excluding the L3, Montecito is roughly a 170 million transistor CPU, 63.7 million if you're just counting logic portions.

By Anonymous Freak on 2/26/2007 10:21:48 PM , Rating: 3
This isn't like making a combo gas/diesel vehicle, this is more like making a vehicle that can have either a gas engine or a diesel engine that are easily exchanged. Presently, (to continue my car analogy,) it is like all diesels being big trucks with completely separate manufacturing than gas trucks. This will be like moving to have one manufacturing line, where at the end, you either put in a gas engine or a diesel engine.

In short, Xeons will still be based on the mainstream desktop processor architectures, but the support structures (chipset, front side bus, etc,) will be common to both Xeon and Itanium. This is more like what happened with Centrino's original 'Pentium M' processor. It was more directly based on Pentium 3 than Pentium 4, but it used the Pentium 4 front side bus rather than the old outdated Pentium 4 FSB.

By Dianoda on 2/27/2007 12:05:02 AM , Rating: 3
Think of this as Intel's way of cutting costs where it can from the good (though troubled is closer to reality) ship Itanium. A convergence of platforms means one less platform for Intel to develop. It makes plenty of business sense assuming it doesn't cause any more problems than the one it solves. Intel knows that Xeon is the priority here, so they won't take any unnecessary risks when it comes to protecting that line of products, but it really would like to get Itanium closer to generating a ROI.

About time
By saratoga on 2/26/2007 9:48:53 PM , Rating: 2
This will be a good thing for Intel. Being able to reuse x86 tech in Itanium will make that platform a lot cheaper to maintain, and greatly slow its gradual disintegration, allowing Intel more time to recoup development costs.

I never understood why they kept delaying it; CSI should have been here a long time ago.

RE: About time
By enumae on 2/26/2007 11:09:12 PM , Rating: 2
Itanium is apparently doing better...

EPIC or Itanium-based systems grew 71.5% year over year, generating more than $1.1 billion in revenue for the quarter the first time the platform has exceeded $1B in revenue in a quarter.

By Mumak on 2/27/2007 4:49:57 AM , Rating: 4
Tigerton (Conroe core) will not be CSI. That will be the Gainestown based on Nehalem core.

But will it run OpenVMS?
By cowzeetgrass on 2/27/2007 8:07:08 PM , Rating: 2
Have to get it out of the way, long time reader, first time poster. (mark me down if/as you see fit)

I'm just curious from a business standpoint, if it will still run OpenVMS. We use it as our main OS mainly because it's a tank as far as operating systems go. I've personally seen it running clean for 3+ years. I think DEC (Digital Equipment Corp) have it documented at 10+ clean. It's not as flashy as most of the operating systems today, but when you're dealing with government agencies, financial institutions, healthcare, etc..., the data doesn't have to be pretty, it just has to be there.

Wiki Links:


"If a man really wants to make a million dollars, the best way would be to start his own religion." -- Scientology founder L. Ron. Hubbard
Related Articles
Intel Responds to AMD's "Torrenza"
September 27, 2006, 9:30 PM
The Count of "Montecito"
July 16, 2006, 3:39 PM

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki