BLAS autotuning

Open discussion for MAGMA library (Matrix Algebra on GPU and Multicore Architectures)

BLAS autotuning

Postby giammy » Wed Jun 23, 2010 8:55 am

Hi all,

Anyone has an idea if there will be some implementation of automatic tuning
(something similar do ATLAS) for GPUs?
Is there (or there will be) any similar mechanism in Magma?

Posts: 1
Joined: Wed Jun 23, 2010 8:50 am

Re: BLAS autotuning

Postby robdhunt » Wed Jun 23, 2010 9:03 am

There has been some work done here:
Posts: 2
Joined: Mon Jun 07, 2010 7:45 am

Re: BLAS autotuning

Postby Stan Tomov » Thu Aug 05, 2010 5:27 pm

Yes, our kernels are "auto-tuned" (not totally automatic yet). Some of the
our work on auto-tuning MAGMA BLAS kernels is described in

Li, Y., Dongarra, J., Tomov, S. " A note on auto-tuning GEMM for GPUs,"
Proc. of ICCS'09, Baton Rouge, LA, UT-CS-09-635, May 25-27, 2009.

Nath, R., Tomov, S., Dongarra, J. "Accelerating GPU Kernels for Dense Linear Algebra,"
Proc. of VECPAR'10, Berkeley, CA, June 22-25, 2010.

We released the sources for our new SGEMM and DGEMMs for Fermi. The algorithms had
to be extended so the search space for the "automatic" search for new algorithms was also
expanded (e.g., this is how the sgemm was derived from the dgemm implementation).
Note that these kernels can be used to speedup most of the other Lavel 3 BLAS by expressing
them in terms of GEMMs, or third party efforts to develop auto-tuned BLAS.
Stan Tomov
Posts: 255
Joined: Fri Aug 21, 2009 10:39 pm

Return to User discussion

Who is online

Users browsing this forum: Bing [Bot] and 0 guests