News Mac Supercomputer: Fast, Cheap

As has already been posted in GN&SD, the 'Big Mac' is a supercomputer made up from a whole load of G5 Macs. Wired has a report on the initial benchmarking of this monster:

Early benchmarks of Virginia Tech's brand new supercomputer -- which is strung together from 1,100 dual-processor Power Mac G5s -- may vault the machine into second place in the rankings of the worlds' fastest supercomputers, second only to Japan's monstrously big and expensive Earth Simulator.

The Big Mac's final score on the Linpack Benchmark won't be officially revealed until Nov. 17, when the rankings of the Top 500 supercomputer sites are made known at the International Supercomputer Conference.

But Jack Dongarra, one of the compilers of a Top 500 list, said Tuesday that preliminary numbers submitted to him suggest Big Mac could be ranked as high as second place.

More here

No matter if you're a Macophile or PC zealot, this is a stunning achievement. I bet ASCI Q (the current no. 2) didn't cost $5m!Just awesome...

 

2nd place ? Amazing .... didnt expect that.

I am not overzealous about mac or PC ... if it gives you a reasonable cost/performance, then use it.

 

1100 duals?

2200 cpus

4400 Ghz, in intel terms, probably equal to 6000ghz or so



bloodey nora, easily second

 

isnt that 6 terahertz?

 

4400 CPUs huh? I'd like to see how that puny apple does against 4400 Pentium 4 3.2EEs OC'd to 4 ghz. Probably end up as applesauce.

 

Slashdot just posted something on the Athlon 64 pwning the G5. I'de like to see a competition of 1100 dual G5's and 1100 dual A64's.

Its still pretty amazing in either case.

 

Yeah, the athlon 64 totally kicks the pitiful g5s ass. Although the EEs would still be faster, it would be niiiice to see apple get smoked either way

 

Oh do shut up. Stop being such a zealot and acknowledge that the Mac is a good machine and that this installation is an ingenious and brilliant idea.

*n

 

its not like it was an original idea...

ok, they used Mac's which was unexpected, but the underlying theory was sound - and not theirs.

 

At this scale and level of processing power I would call it ingenious

Especially if it grabs that #2 spot.

*n

 

its certainly an achievment to get all that working, to even have the brass to consider such a risky venture is something. I just wouldnt call it ingenius as I havent seen them do anything 'new'. Only the collaboration of previous attempts on smaller scale.

Still, to be #2 super-dooper-computer, must be a boost for both the university & apple.