[Beowulf] I'm so glad I didn't buy one of these

Prentice Bisbal prentice.bisbal at rutgers.edu
Thu Jul 3 07:45:52 PDT 2014

On 07/02/2014 05:37 PM, James Cuff wrote:
> Let this be a lesson to us all.
> Repeat after me:
> "Top 500 scores mean nothing!"
> As a community we have to stop this madness!  100k / day.  Sigh.
> J.

There is signs of this madness ending, at least in the US. I read a few 
studies about how computational scientists work, and they all came to 
the same conclusion: Most computational scientists cannot use, or have a 
need for large supercomputers. Most just need a decent departmental 
cluster or institution-wide cluster. I can't find links the specific 
papers I read, but I believe I found them on this page of Software 
Carpenty site:


The NSF has sent out RFPs (like NSF 13-528) that aren't for 
LINPACK-killing supers, but for more 'accessible' systems. Things are 
slowly changing, but just like Formula 1 and NASCAR, people are going to 
follow the fastest. Last I checked, there weren't huge crowds at my 
local autocross races. :(


More information about the Beowulf mailing list