[Trilinos-Users] CUDA (not trilinos related)

[Baker, Christopher Grover]

The story seems to be different for sparse mat-vecs.

A somewhat recent (Dec 2008) report out of NVIDIA shows significant speedup of CUDA-based sparse matvec, in single and double precision, over both Cell and multi-core CPU.
They distribute code and matrices along with the report.

See:
http://www.nvidia.com/object/nvidia_research_pub_001.html

Chris


On 7/29/09 2:38 PM, "Matt G" <mgoodman at email.arizona.edu> wrote:

This is true as long as the multiplications performed don't need to exit the scalar processors.  For most realistic linear algebra tasks you bump into the memory bandwidth limit way before reaching that.  You only ever get the theroetical performace when doing stuff like fractals <http://scipyed.wordpress.com/2009/01/03/pycuda-fun/>  or wpa cracking <http://hothardware.com/News/Russian-Firm-Uses-NVIDIA-GPUs-To-Crack-WPA-WPA2/> .  Benchmarking the cublas implementation vs. cpu blas for DP typically gives a win to the CPU.  I have yet to have mkl loose to a gpu (295's included), admittledy I am rolling with a pricey i7.

All the magic in cpu blas comes from consecutive cache hits and never needing to wait on memory.

-Matt

On Wed, Jul 29, 2009 at 12:09 PM,  <rrossi at cimne.upc.edu> wrote:





A comment on CUDA ... which indeed has very little to do with trilinos.

concerning the on-going discussion on CUDA performance, i would like to say that Matrix-Vector Multiplication using nvidia GTX280 and DOUBLE PRECISION is at 5 to 10 times faster than a recent quad core system (old quad core ... not sure of the I7).

If the matrix is sufficiently big this figure could be even better.

I should say that the figures i report are obtained using a serial code for the matrix multiplication based on the ublas library. I am pretty sure in any case that even trilinos is not faster than that for a single CPU.

greetings
Riccardo

_______________________________________________
Trilinos-Users mailing list
Trilinos-Users at software.sandia.gov
http://software.sandia.gov/mailman/listinfo/trilinos-users



-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://software.sandia.gov/pipermail/trilinos-users/attachments/20090729/43385f26/attachment.html