Slashdot Mirror
IBM Kills project Monterey
I just got this news - IBM is killing project Monterey. Full story can be found on this page at ZDNET (Smart Partner). This is a bit surprising (if I may call it like this).
← Back to Stories (view on slashdot.org)
Monterey is very much alive. The only thing that is going on is that the RELEASE name is changing from Monterey/64 (on IA64) to AIX 5.0 (or AIX-L, with the "L" standing for the ability to run Linux IA64 binaries). SCO will probably ship "AIX 5" for IA64. It is NOT the case that IBM is going to be using the Linux kernel. One thing that was supposed to happen that might get dropped is support for UnixWare IA32 binaries on IA64.
No, article should read: IBM notices Itanium is dead.
Heads up, folks: Itanium is never going to come out in volume. It's years late and horribly underperforming.
The design was finished over two years ago, but it took Intel over a year to go from finished design to tapeout. This indicates that the original design ran into huge problems and required many many revisions when they tried to translate it to actual silicon. The total on-die cache of Itanium is paltry a 128kB, less even than the crippled cache on a Celeron; this indicates that the core logic of the chip took up much much more die space than they planned, and they only had enough room for a token cache. (As the chip is an in-order design and its primary functions involve high data throughput, it needs more low-latency cache than an out-of-order chip, not less.) As recently as 2 months ago Intel was claiming they'd be selling Itanium systems by now. That would mean volume production of 800 MHz and higher chips. At Linuxworld last week, only 1 of the demonstrated systems had chips running over 500 MHz. That means Intel is having some serious, serious problems with their fabbing--they can't even get a couple demonstration chips running at 800 MHz, much less volume production. Think about that for a second--Itanium is supposed to be Intel's new flagship server chip, yet they can't even give it as much on-die cache or even as high clock speeds as their slowest Celeron?? Talk about embarrassing.
All these signs, plus Intel's ever-increasing delays on the chip, point to the fact that Itanium ran into design problems so serious that the project is in fact dead. Since they've built up so much hype behind it, though, Intel will keep announcing delays until they can finally claim that the reason they're not debuting Itanium in volume is because McKinley is just around the corner. (Expect a McKinley paper launch in 2H '01, with volume by the beginning of '02.) Indeed, some rather credible rumors eminating out of Intel now include that 1) Itanium is yielding so poorly that a relayout and critical path trim will be needed before it could possibly yield 800 MHz in volume; 2) the current versions are suffering from an errata in the power management circuitry which is limiting their clock speeds; and 3) Itanium is in fact yielding ok, but the extraordinarily complex compilers needed for EPIC are currently producing such slow code that Intel needs to pretend they can't make the chips to save face. Obviously all three aren't true, but the chance that one is is pretty high.
Meanwhile, McKinley (which, not so coincidentally, has been designed almost entirely by HP) looks on target to hit its rather impressive performance objectives, if a little late. MS--whether by choice or necessity--has decided to wait for McKinley before releasing NT-64, rendering Itanium pretty much dead anyways. If the compilers turn out ok, McKinley will probably be able to compete on a performance basis with Sun's US3, and maybe even IBM's Power4. (The Alpha 21364 will cream it, but what else is new?) Meanwhile, Itanic looks to be slower than a Celeron for integer code and about even with a high-end P3 for fp; the P4 if not the GHz P3 ought to beat it handily even in the rather server-oriented SPEC suite. Having your $3000 server chip beat in SPECfp by your mainstream desktop chip is an embarrassment Intel will never allow, even if that means not releasing Itanium at all.
Who would buy such a chip? Almost no one. The only Itanium boxes ever sold will be to ISV's who need development platforms for McKinley, and large corporations who want to prepare for McKinley. And they can all be served by development systems and engineering samples; there's no need for volume production of a chip that's only going to be used for development and validation.
And that's why this announcement is not a surprise. IBM accomplishes four things with this: 1) they associate themselves more with Linux, which is becoming a bigger part of their strategy these days; 2) they disassociate themselves from the Itanic, which is a bit of an industry joke; 3) they shift the emphasis to their upcoming Power4 architecture, which looks to be quite good and a worthy competitor to anything IA-64 will produce for a while; and 4) they still keep IA-64 compatability for the future, albeit less emphasized from a PR point of view.
All in all a smart decision. And not a surprising one.
This is not actually that surprising. IBM has a stated goal of making linux run on every piece of hardware and every platform they sell, from the top of the line (OS390) to the bottom (intel-based netfinity line, I guess).
So there are two things going on here: 1) IBM has their own version of Unix that's quite good but not doing very well in the marketplace. 2) IBM has decided that Linux is the way to unseat Sun's dominance of the midrange server market.
Given those two facts, supporting yet a different version of Unix designed primarily for the Itanium platform (regardless of what they say about also running on the Power chip) doesn't make any sense. Even IBM has limited resources.
Bullshit. The purpose of OSF was to unify against the looming threat of SunOS/AT&T SysV integration - it would do excatly the opposite of protecting "its members' respective proprietary OSes." Sun eventually parted ways with AT&T, and OSF withered. DEC was the only company to actually release OSF for its hardware. IBM and HP eventually went with a SysV strategy anyway; Digital/Tru64 unix remains the only commercial unix that is largely based on the BSD code.
--
http://gammatron.weblogger.com
Monterey was the joint effort of IBM, SCO, and two others to port a high-end, enterprise class unix to Itanium (IA-64). The excitement driving the buzz was that Monterey looked like the migration path for AIX.
Looks like Linux inherits all that buzz.
GO TRILLIAN
"one treats others with courtesy not because they are gentlemen or gentlewomen, but because you are" --G. Henrichs
I amazed that they even ever saw SCO as a viable partner - the corpse of SCO has been floating from door to door looking for some poor sucker to take it in and break it down for spare parts. Caldera finally was suckered. Ransom Love looked quite clueless telling the audience in San Jose that Linux alone couldn't do it - that somehow SCO's dead product line was needed to complete its promise to customers. What hooey. SCO will be like WordPerfect, a forgotten power that drifts from buyer to buyer. Caldera needs to realize that customers want to hear a coherent marketing story - having a linux company come out and tell people that linux is inadequate is not what I would call a compelling marketing story. This doesn't surprise me one bit - Caldera has never known one thing about marketing their own product (the "Business Linux"???? what the hell is that????).
Anyway, back to IBM. Its nice to see finally that the potential market for AIX on IA64 is likely too small to address as a strategic issue. Customers are tired of parallel product lines that somehow address high-end, midrange, low-end, in some bizarre drivel that never makes any sense. Look at Compaq's worthless unix marketing plan regarding Tru64 and Linux.
IBMs major problem has always been that it considers itself too big to commit to any one platform. This is why still to this day IBM has marketing issues. Look at Sun - they have one product line and one OS - a Sun customer always knows what end is up with that company and the Sun commitment is always coherent. This is why Sun is going to continue being the number one unix vendor, for better or for worse, even though IBM's product lineup is likely superior (just impossible to see in continuity).