• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Pics of Virginia Tech's G5 Cluster

Barnaby W. Füi

Elite Member
Pics here: http://don.cc.vt.edu/

From slashdot: "The Virginia Tech G5 cluster has achieved around 80% of its peak performance in preliminary Linpack testing with 128 nodes according to Jack Dongarra at the Top 500. "They're getting about 80 percent of the theoretical peak," Dongarra said. "If it holds, and it's unclear if it will, it has the potential to be the world's second most powerful machine." Typically getting 60% of peak in the Top 500 lists is quite good. If the Big Mac cluster achieves 60% of peak it would displace the 2,300 2.4 GHz Xeon cluster at LLNL for the number three spot on the current list."

http://apple.slashdot.org/article.pl?sid=03/10/16/1814254&mode=thread&tid=126&tid=181
http://www.wired.com/news/technology/0,1282,60821,00.html/wn_ascii
 
Originally posted by: BingBongWongFooey
Pics here: http://don.cc.vt.edu/

From slashdot: "The Virginia Tech G5 cluster has achieved around 80% of its peak performance in preliminary Linpack testing with 128 nodes according to Jack Dongarra at the Top 500. "They're getting about 80 percent of the theoretical peak," Dongarra said. "If it holds, and it's unclear if it will, it has the potential to be the world's second most powerful machine." Typically getting 60% of peak in the Top 500 lists is quite good. If the Big Mac cluster achieves 60% of peak it would displace the 2,300 2.4 GHz Xeon cluster at LLNL for the number three spot on the current list."

http://apple.slashdot.org/article.pl?sid=03/10/16/1814254&mode=thread&tid=126&tid=181
http://www.wired.com/news/technology/0,1282,60821,00.html/wn_ascii

I was going to post this tonight but I went to the shop and forgot!
 
Originally posted by: BingBongWongFooey
Yeah, those suckers put out a lot of heat, and the case is specifically designed to take care of it well.

I think Apple is working on the heat problem for their G5 notebooks. The notebooks will probably coincide with a G5 XScale (rackmount).
 
Originally posted by: XBoxLPU
Originally posted by: BingBongWongFooey
Yeah, those suckers put out a lot of heat, and the case is specifically designed to take care of it well.

9 fans correct ?

Hm, I don't recall the exact number, but I do remember it being odd, and 9 seems about right. Pretty sure it's 9.
 
The case is composed of nine-temperature controlled fans in different "climate zones". For example, Mac OS X might inrease the fan speed in the Processor/Memory Zone if needed more than it probably would in the PCI Zone.

As for the choice of system the benchmarks across the internet show that the Opteron and Athlon64 clearly defeat the G5. Not to mention you could fit more processors in 1U cabinets in there and probably cost about the same as all of those G5s. Unless they have software specifically optimized for the Altivec Engine.

-Por
 
I live only about 2 hours away from Va Tech. I wonder if they'd give me a tour? I could take some photos of my own to post up.
 
Actually the G5 cluster was pretty cheap-

The machine is the first supercomputer based on Macs; it is one of the few supercomputers built entirely from off-the-shelf components and it cost a bargain-bucket price -- only $5.2 million. By comparison, most of the top 10 supercomputers cost about $40 million and up. The Earth Simulator cost $350 million.
 
Originally posted by: beatle
Too cool! 😀 I wish I was still a student there... I'd love to see it up close. :Q

I AM a student here at VT.... and they won't let you get too close.... plus, they're working on it DAILY. 🙂

As far as the nine fans per case goes--yes that's true.... BUT, that doesnt' help you get the heat out of the room. They've installed new A/C units specifically for the supercomputer.

Its also not available for research use until JANUARY. They're apparently having stability problems. Also, I've heard that actual estimates say they'll be able to get 60% of theoretical when all 1100 nodes are up. I'm looking forward to using it--i'm tired of waiting a week to get results as it is. The supercomputer won't help that much, speed-wise..... but it will let me run several concurrent runs (which is always nice).
 
Back
Top