Google's cluster

Ken Chase math at velocet.ca
Sat Mar 15 23:23:40 PST 2003


On Sat, Mar 15, 2003 at 10:06:08PM -0600, Brian D. Ropers-Huilman's all...
> On Sat, 15 Mar 2003, Ken Chase wrote:
> > 
> > As for academic clusters, I expect that more than most grant money isnt handed
> > out around the world just to get onto the top 500 cluster list, but in fact,
> > to get some real work done.
> > 
> 
> While I have already said it before, it still pains me to say it again. Our 
> cluster, currently at #17 on the Top 500 and the 3rd fastest (benchmarked) 
> Linux cluster (we were #11 and the #1 Linux cluster until the list was 
> officially published), was purchased soley to obtain a spot in the top 10 on 
> the Top 500 list. Period. The science that would be run on the cluster was 
> secondary. The governor of the state made it clear that the money he was 
> granting needed to "show something." A spot in the Top 10 was the goal.

Curious: if you had spent that money to do the science you are doing right now
more efficently, what multiple in total throughput of jobs per month would you
now be dealing with over your current cluster?  And what ranking would you
have been in the top 500 in that case?

/kc

> 
> --  
> Brian D. Ropers-Huilman                        (225) 578-0461 (V)
> Systems Administrator                 AIX      (225) 578-6400 (F)
> Office of Computing Services       GNU Linux   brian at ropers-huilman.net
> High Performance Computing            .^.      http://www.ropers-huilman.net/
> Fred Frey Building, Rm. 201, E-1Q     /V\                          \o/
> Louisiana State University           (/ \)           --  __o   /    |
> Baton Rouge, LA 70803-1900           (   )          --- `\<,  /    `\\,
>                                      ^^-^^              O/ O /     O/ O

-- 
Ken Chase, math at velocet.ca  *  Velocet Communications Inc.  *  Toronto, Canada
Wiznet Velocet DSL Datavaults 24/7 help: 416-967-4414 tollfree: 1-866-353-0363




More information about the Beowulf mailing list