[Beowulf] New HPCC results and the Myri viewpoint

Brian D. Ropers-Huilman bropers at cct.lsu.edu
Thu Jul 21 09:43:57 PDT 2005

Hash: RIPEMD160

Timothy Bole said the following on 2005.07.20 23:55:
> A Beowulf is engineered to
> solve a problem, not the other way around.

I think we'd all agree that this is the way it should be, but I contend it
is rarely the case. I certainly don't have the luxury of designing systems
to match a problem, though I so wish I could. Rather, I have a very generic
resource available to a multitude of researchers with many and varied problems.

In this case, I _encourage_ our users, to the extent that is possible, to
optimize their code for our platform.

Likewise, our applications framework, Cactus, is heavily optimized for
various different systems, so no matter where it runs, its going to run as
optimally as possible.

Picking a nit, of sorts, but I think it's an important point to make.

- --
Brian D. Ropers-Huilman  .:. Asst. Director .:.  HPC and Computation
Center for Computation & Technology (CCT)        bropers at cct.lsu.edu
Johnston Hall, Rm. 350                           +1 225.578.3272 (V)
Louisiana State University                       +1 225.578.5362 (F)
Baton Rouge, LA 70803-1900  USA              http://www.cct.lsu.edu/
Version: GnuPG v1.2.4 (Darwin)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org


More information about the Beowulf mailing list