This is not for design of jets but, booster rockets to lift payloads.
This very small cluster(96x) is for CFD, computational fluid dynamics.
For the light duty AMD is ok and cheap since SMP is not required. CFD on booster rockets has been around for the past 30+ years. step into GFD (geophysical fluid dynamics) and AMD is no where to be found.
Its just a monumental point for AMD actually scoring a cluster bid, boeing went the cheap route. nothing wrong with boeing saving a buck when computational power vs cost is a factor (since we been putting rockets in orbit for 30+ years). Amazing what sales people can sell these days.
Awe, just because businesses are finally waking up to the performance benefits of an Athlon based system, in addition to the cost advantages, is no reason to get snipy. Or are you suggesting a 96 processor solution from Intel would beat it, costs be damned? I don't think even you would be that stupid, despite countless posts suggesting otherwise.
why does everyone in the forum always have to "one up" everyone else? does it make you guys feel better to be able to put down someone you dont even know? most people like those bickering in this post try to act all high and mighty, arguing over the stupidest things ad finitum. please cut your comments down to a purer state and leave out those amd or intel directed barbs.
You've got to be kidding me!!! Do you honestly think Boeing, **BOEING**, would select AMD becuase they saved a few dollars? FUGGER, does Intel pay you to spew your worthless propoganda? If not, what motivates you? I can't even begin to understand why there are Intel droids like yourself who are so devoted to defending a company that doesn't even know you exist.
This is not the first cluster AMD has won. I recall reading an article several months ago about another AMD cluster.
There is a pretty major flaw in your 'cheap' comment. In business, if a department needs to buy some hardware, they send a capital expenditure request to their corporate uppers. They look at the request, putter around with the numbers, and respond with how much they can actually spend. So in this case, Boeing set aside X dollars to spend on a cluster to analyze data. How would spending that money on an ellegedly inferior cluster be the more cost concious decision? If an Intel solution is better, why didn't they buy it?
Just for the record, IBM announced yesterday that it will be releasing Dual processor Athlon servers. While they may not be available yet, I think that is a pretty good indication that the product is both coming to market and is a viable, stable solution. But, I suppose IBM has just been suckered by all the AMD marketing hype and the Lemming mentality it has created. After all, AMD has a HUGE marketing presence (NOT).
Actually Fugger, SMP is often not a consideration at all for clusters. It all depends on your application, but if you need lots of memory, memory bandwidth, and network bandwidth per CPU, you never want more then 1 CPU per motherboard. A CFD problem fits this desciption very well due to the very large matrix sizes involved.
If you have smaller individual tasks, and stuff that can be efficiently threaded (as apposed to message passing), SMP is great and can significantly reduce the size & cost of a cluster. I may have to go with an Intel solution just because it can fit in one rack instead of two, despite the performance hit :-(
And by the way, trans-sonic CFD is NOT an "extreme light duty" application by any means.
My guess is Boeing went with AMD because they could get by with 96 CPUs instead of 120 Intel CPUs, and at a lower cost. In fact, this is exactly what Linux Networx reccomends for floating point compute bound applications. I know, because I was just discussing a 32 node cluster with them not 5 minutes ago. They promised to send some AMD vs. Intel floating point benchmarks with the quote.
In theory, there is no difference between theory and practice.
In practice, there is.
You wrote: "The reason this article stand out is that AMD and cluster were never used in same sentance before. Its really a joke since the extreme light duty of this CFD model requires"
Well, it appears they are begining to be mentioned more often. See this link http://www.tu-chemnitz.de/mbv/TechnThDyn/mpf/mpf_athlon_cluster.html
They do not seem to think that their application requires "lightweight"duty.
And This link:http://biz.yahoo.com/bw/010122/ca_amd_3.html
for info on the Beowulf cluster.
Dig at all the angles you want, Ill tell you the same.
I wont waste my time. no copy, no paste, no search for URL. sorry your on your own. stay clueless.
Im not about to open myself up to your lame ass critique of my reply on SMP.
Sojourn, bieng as we do not know what boeing budgeted for this project lets not assume. but we can look at Linux NetworX and see that they are "cheapest" cluster solution provider available (just short of in house techs building custom). so price was a factor.
Boeing a big company going cheap route...
One thing we know for sure, SMP was not a factor.
If SMP were a factor, AMD would have not been quoted in this bid.