Re: [boost] uBLAS formal review

From: Martin Weiser (weiser_at_[hidden])
Date: 2002-06-28 04:37:17

On Donnerstag, 27. Juni 2002 22:25, Peter Schmitteckert (boost) wrote:
> Thanks for doing this graph, I was planning to do this kind at the
> weekend, so you saved my weekend :)

You're welcome.

> In my application the difference between plain ublas (gcc-2.95.4,
> amd/athlon, hp pa-2.0) and a version where most ublas-prod expressions
> are replaced by ATLAS dgemm (vendor blas for hp) gives an overall
> improvement of a factor 5 to 10.  The matrix sizes are ~100x100 and
> below. So, the use of dgemm is a must. Using xgemm instead of prod()
> for matrices which are suited for xgemm is fairly simple (using
> #ifdef).

Even though it's fairly easy, a specialized version of prod() calling 
xgemm under the hood would be even easier to use. Insofar the interface 
is influenced by the ability to specialize core operations, which is why 
I started this thread during review in the first place.

Yours,
Martin

-- 
Dr. Martin Weiser            Zuse Institute Berlin
weiser_at_[hidden]                Scientific Computing
http://www.zib.de/weiser     Numerical Analysis and Modelling