Itanium Retreats To Multis, Opteron Presses Attack

vincecate writes " Back in 1997 Itanium was presented as a compatible extension of x86, and Intel had dreams of replacing high end x86 by 1999 (about min 16 in link). However, the original 2001 Itanium had the 32-bit performance of a much older Pentium so Itanium is viewed as non-compatible. Now with the rise of what Microsoft is calling the "x64" architecture (AMD64 and EM64T), Itanium has been repositioned to the multiprocessor server market."

However, Opterons are designed for multiprocessing. Each has its own memory and built in hypertransport links to make NUMA multiprocessors. So existing Opterons don't need large expensive caches like Xeon MP, nor any glue chips for up to 8-way multiprocessors. This has started the commoditization of multiprocessors. You can now buy a 4-way Opteron for $6000 or an 8-way for $10,000.

There is an interesting interview with Tyan CEO and his assistant that give some info about future Opteron plans. People are working on 8-way motherboards that will become 16-way when duel-core Opterons come out. They also say that the next Opteron core, the K9, will be able to 'go over 60 processors without adding any external crossbar chips.' Another fun plan mentioned is connecting the hypertransport links using fibers for really large systems."

Comments

Intel attempts to move away from x86

[vincecate]

The AMD multiprocessor plan seems like an Itanic killer. History will probably record Itanium as Intel's VLIW attempt to break away from x86. Very similar to their failed IAPX_432 object oriented attempt and their i860 RISC attempt to move customers away from x86. Each time Intel seemed to use the architectural fad of the day to try to move people off x86 but without good backward compatibility. Each time seems to have failed badly.

Re:Intel attempts to move away from x86

[questionlp]

Although AMD is not directly aiming the Opteron at the Itanium (instead it's aimed more towards the Xeon DP and Xeon MP), the Opteron can be very competitive against the Itanium in a fair number of scenarios. There are some places where the Itanium will beat the AMD in performance, but it also costs more. For instance, the Itanium (with a bit of help from SGI) can scale into the hundreads of processors and still be able to run a single kernel image. Right now, that isn't exactly doable with the Opteron.

Intel is aiming their Xeon with EM64T at the Opteron while pinning the Itanium against Sun's UltraSPARC, IBM's POWER, etc. Unfortunately, the Itanium is also used as a replacement for the PA-RISC and the Alpha :\

I favor the Opteron over the Itanium because of backwards compatibility without performance problems, lower cost, lower power consumption and better I/O and memory throughput.

I have bashed the Itanium in the past... but I still think it is a bit too early to sent it crashing towards an iceberg and sinking it soon. It does have it's good qualities, but also a fair amount of downsides.

US$6000/1000 for 4/8 CPUs

[hbackert]

Those offerings from Rocketcalc are dual Operon boards with orginary clustering. The only difference is, they put them in one case. A 4 CPU or 8 CPU Opteron box is far more expensive. of course a 8 CPU Itanium2 is expensive too. But comparing a bunch of (commodity) dual CPU boards with one 8 CPU box is not fair. It's the often found apple-and-oranges-comparison comparison.

Re:US$6000/1000 for 4/8 CPUs

[FueledByRamen]

Last I checked, Tyan's 4way Opteron boards start at $1500 and change. And then you have to get the 8xx series processors. Not exactly cheap - $1250 for the Opteron 848, $855 for the 846 (or if you really have cash to blow, $1549 for the 850). And then you need some good registered ECC DDR400 RAM - $120 per 512 meg stick of Kingston, and I'd want to put several gigs in a machine like that... Plus incidentals like a huge case, really big fans and an even bigger power supply, and you're getting into some serious cash outlay - $8000+ for a 4x 848 with 4 GB of RAM.