Fortran compilers for Linux/mpich

Robert G. Brown rgb at phy.duke.edu
Sun Nov 25 13:22:16 EST 2001


On Sun, 25 Nov 2001, David van der Spoel wrote:

> On Sun, 25 Nov 2001, Robert G. Brown wrote:
> 
> >Do you have any idea what the overall marginal benefit is of using your
> >hand-optimized routines when working on large datasets (too big to fit
> >into cache)?  In particular, does performance devolve to
> >memory-bandwidth-bound behavior (and hence end up being the same for
> >MILC and SSE and dominated by the memory bus speed)?
> >
> >    rgb
> Of course YMMV, but for our application (molecular dynamics) the impact of
> SSE is high: a factor of 1.5 for large applications, more than so for
> smaller applications (see http://www.gromacs.org/benchmarks/scaling.php
> for comparisons). I should admit that it was very time consuming to write
> all that much assembly code (but the guy did it out of his own free will)

I've been meaning to go back and play with this -- there must be some
way of quantifying the crossover point between CPU bound and memory I/O
bound code, and I've got a decent benchmark timing harness at this point
that I can use to explore it.

It's good to hear that it can yield a real benefit for large data codes
though.

    rgb

> 
> 
> Groeten, David.
> ________________________________________________________________________
> Dr. David van der Spoel, 	Biomedical center, Dept. of Biochemistry
> Husargatan 3, Box 576,  	75123 Uppsala, Sweden
> phone:	46 18 471 4205		fax: 46 18 511 755
> spoel at xray.bmc.uu.se	spoel at gromacs.org   http://zorn.bmc.uu.se/~spoel
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> 
> 

-- 
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



More information about the Beowulf mailing list