3000+ cpu G5 Cluster?
Other than Virginia Tech.
Heard whispers about this, or something similar from 3 different places now, the latest of which is over at MacNN...
http://forums.macnn.com/showthread.p...88#post2045188
Anyone know anything about it over here?
Heard whispers about this, or something similar from 3 different places now, the latest of which is over at MacNN...
http://forums.macnn.com/showthread.p...88#post2045188
Anyone know anything about it over here?
Comments
http://maccentral.macworld.com/news/...=1087832245000
Apple Computer Inc. will announce on Monday the sale of 1566 dual processor 1U rack-mount 64-bit Xserve G5 servers to COLSA Corp., which will be used to build what is expected to be one of the fastest supercomputers in the world. The US$5.8 million cluster will be used to model the complex aero-thermodynamics of hypersonic flight for the U.S. Army.
"We did about a year and a half of research on a variety of processors before making our decision," Dr Anthony DiRienzo, executive vice president at COLSA Corp., told MacCentral. "We did a best value competition and Apple won that competition. It was based on performance; the facility (power requirements, floor space etc.); cost; and an assessment of vendor stability. We solicited to six companies and they won."
The supercomputer, named MACH 5, is expected to deliver peak performance capability of more than 25 TFlops/second. In comparison, the Virginia Tech supercomputer announced last year attained sustained performance of approximately 10 TFlops/second, according to Apple director of product management, server hardware, Alex Grossman.
With those numbers, the MACH 5 would rank second only to Japan's $350 million Earth Simulator computer.
"We evaluated PC-based proposals from other vendors but none came close to delivering either the price, performance or manageability of the AppleXserve G5," said DiRienzo.
The Xserve G5 supercluster system is expected to be online and working for the Aviation and Missile Research, Development and Engineering Center (AMRDEC) division of the US Army Research and Development Command by late Fall.
Shortly, DiRienzo said they would take delivery of 300 Xserves a day, set them up in the racks and the next day begin the process again until all of the Xserves are installed and working.
as
2200 2GHz G5s ~ 10 TFLOPS
3132 G5s ~ 25 TFLOPS
Originally posted by Existence
Can anyone tell me what information we're missing?
2200 2GHz G5s ~ 10 TFLOPS
3132 G5s ~ 25 TFLOPS
Maccentral sucked at that part of reporting.
Big Mac has sustained power of 10 TFlops. Mach 5 has hopefully a peak of 25 Tflops, but hopefully a sustained of 15 TFlops.
Originally posted by Existence
Can anyone tell me what information we're missing?
2200 2GHz G5s ~ 10 TFLOPS
3132 G5s ~ 25 TFLOPS
2200 2Ghz G5 in PowerMacs
3132 2Ghz G5 in XServes
Originally posted by Existence
Can anyone tell me what information we're missing?
2200 2GHz G5s ~ 10 TFLOPS
3132 G5s ~ 25 TFLOPS
It's actually:
2200 2 GHz G5s = 17.6 TFLOPS
3132 2 GHz G5s ~ 25 TFLOPS
Those are the peak figures for each. The sustained numbers were ~ 10 TFLOPs for Big Mac and not yet known for the latter.
Still, kudos to Apple for making this sale!
Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.
Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
Originally posted by Telomar
It's actually:
2200 2 GHz G5s = 17.6 TFLOPS
3132 2 GHz G5s ~ 25 TFLOPS
Those are the peak figures for each. The sustained numbers were ~ 10 TFLOPs for Big Mac and not yet known for the latter.
Figure around 55% efficiency. The Army cluster will be not be using Infiniband fabric (as VT did), so I estimate they'll get somewhere around 12.5-14.0 TF Max. They're probably shooting for 15TF a second.
dur, they are shooting for "one of the" fastest.
eitherway, if a couple more institutions make their own 'big macs' then apple products could easily fill up most of hte top 10
Originally posted by Tomb of the Unknown
Figure around 55% efficiency. The Army cluster will be not be using Infiniband fabric (as VT did), so I estimate they'll get somewhere around 12.5-14.0 TF Max. They're probably shooting for 15TF a second.
Perhaps on linpack. The interview suggested that they didn't go for a hardcore network system because their workloads were very compute intensive, but didn't involve particularly big data sets.
Not sure how much difference it will make, other than "some."
Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.
Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
Originally posted by mooseman
...the two new ones should probably have slightly higher efficiencies, given that no units will have to be used for error correction like they were with the Big Mac cluster. The XServes use ECC RAM, where the plain ol' G5 Power Mac does not.
I think VT sort of cheated and didn't count the error-checking overhead (which was 2x).