Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
7107 Discussions

multithreaded performance of sparse blas not good

ashade
Beginner
‎04-09-2011 02:56 PM
409 Views

Hi all

I'm having trouble with matrix multiplication on Sparse Blas. I am trying to multiply 2 huge matrices and compare multithreaded and single threaded performance on an quad core AMD Phenom II 940 with 4GB of DDR3 RAM.

I am using mkl_scsrmm. On benchmark, I repeat the call to mkl_scsrmm a hundred times and I compute the total time in seconds. The matrices have 700,000 (dense) and 20,000 (sparse) elements.

The problem is that the multithreaded performance is only 5% better than single threaded sequential performance.

What is happening?

Ashade

0 Kudos

Link Copied

2 Replies
mecej4
Honored Contributor III
‎04-10-2011 05:09 AM
409 Views
With such benchmarks, the devil is in the details. You may be interested to see what Amdahl's Law states regarding your runs.

If you want more specific help, you will have to document what you did, and state reasons why the results are not what was expected.
0 Kudos
Copy link
Gennady_F_Intel
Moderator
‎04-10-2011 08:56 AM
409 Views
Ashade,
Do you mean that you see 5% of scalability with 4 cores vs 1 core?
what is the matrixes size?
0 Kudos
Copy link
Reply

Community support is provided Monday to Friday. Other contact methods are available here.

Intel does not verify all solutions, including but not limited to any file transfers that may appear in this community. Accordingly, Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade.

For more complete information about compiler optimizations, see our Optimization Notice.