[Beowulf] RE: Compare and contrast MPI implementations

[Robert G. Brown]

On Wed, 21 Dec 2005, John Hearns wrote:

> On Tue, 2005-12-20 at 14:52 -0800, Greg Lindahl wrote:
>> Our artificial-stupidity "pathopt" utility can do a pretty good job of
>> magically finding the right compiler flags, but it's not that fast at
>> producing the answer.
> Well, you need to run it on one of these Beowulf clusters to make it faster.

Naaaa, you just need to recursively recompile the pathopt utility with
pathopt a few times until it converges as really really fast...;-)

    rgb

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu