linpack/lapack benchmarking

Byron L. Sonne blsonne-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org
Mon Oct 4 03:27:45 UTC 2004


Greetings,

After spending the evening installing various fortran compilers, tools 
and fortran related bits (Blas, etc.). I've yet to be able to compile or 
run a single usable linpack (http://www.top500.org/lists/linpack.php) 
test. I've tried it from source, and I've tried it from binaries. I've 
looked on Freshmeat and I've been to top500.org, netlib.org, etc and 
grabbed code from there. Not alot of progress.

I should mention right here, right now that the only experience I have 
with Fortran is that I know how to spell it, and that it sounds kinda 
funny pronouncing it backwards.

Anyone have some binaries, code or better yet, a package/tgz that will 
work/compile/actually run on a fairly up to date system (Pentium 4)? Can 
someone tell me how the heck I can accurately calculate the MFlops on my 
box, or show me in the proverbial '10 easy steps' how I can use one of 
these mathemtical frameworks + libs to load up and make the calculation? 
That would be way cool of you. I would really like to have a good idea 
about the various systems in my possesion and how they perform "solving 
a dense system of linear equations".

Regards,
Byron

--
The Toronto Linux Users Group.      Meetings: http://tlug.ss.org
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml





More information about the Legacy mailing list