Re: OSKI, an automatically tuned sparse kernel library.

[Andy Adler]

On Wed, 29 Jun 2005, Richard Vuduc wrote:

> David Bateman wrote:
> > It looks interesting. Octave at the moment has a pretty good sparse
> > matrix mul code, that is significantly faster than matlab for most
> > densities. Matlab beats octave for very low densities.
>
> This is very interesting, and is the sort of issue OSKI addresses
> since tuning is both matrix- and machine-specific.

I've been trying to work on the octave sparse matrix multiply code;
I have a patch for low density sparse matrices that should
increase the performance by about 4x. When I get back from my
vacation (end-July), I'll test and submit it.

I wanted to try to compare this to the OSKI performance. OSKI
provides a "make benchmarks" function with looks promising.
Unfortunately, I couldn't find a way to compare to octave
performance easily.

1. OSKI benchmarks creates its own test matrices. There doesn't
   seem to be a way to enter or save these to be tested elsewhere.

2. I wasn't able to find test code for sparse x sparse function.

Richard, would you be able to offer some advice?

--
Andy Adler <address@hidden> 1(613)562-5800x6218