[Beowulf] Re: Re: Home beowulf - NIC latencies

[Camm Maguire]

Greetings!

"Robert G. Brown" <rgb at phy.duke.edu> writes:

> But maybe this is all too complicated, or doesn't belong in the standard
> per se.  It is indeed like the ATLAS thing, but then, I think that ATLAS
> is sheer genius although it is also cumbersome and clunky to build...;-)
> I just dream of the day that ATLAS-like runtime optimization isn't so
> clunky and is based on tools that create tables of microbenchmark
> numbers that ARE sufficiently accurate and rich to achieve
> near-optimization without running a build loop that sweeps and searches
> a high-dimensional space...:-)
> 

I'm of the opinion that the bulk of the benefit can be had by
providing alternate binaries tuned for the coarse grained cpu
differentials - isa extension, cache, etc. -- coupled with some smarts
in ld.so to select the proper version at runtime depending on the
running cpu.  We do something like this, (alternate isa extension
only), in the Debian atlas packages, which are now at the base of
quite a large application tree -- no recompilation required.

Am always appreciative of other thoughts on these matters.

Take care,


>   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
> 
> 
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
> 
> 
> 

-- 
Camm Maguire			     			camm at enhanced.com
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah