Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
6982 Discussions

Pardiso solver no threaded speedup

Dave17
Beginner
‎03-15-2024 09:24 AM
199 Views

We find that linking the threaded MKL Pardiso solver provides no performance boost - or even slows down if many threads are specified via OPNOMP_NUM_THREADS.

The Pardiso diagnostic messages report multiple threads are being used.  But the execution time slows.

Is this to be expected?

 

Labels (1)
Labels:
0 Kudos

Link Copied

1 Reply
Gennady_F_Intel
Moderator
‎03-19-2024 03:17 AM
144 Views

First of all try to play with MKL_NUM_THREADS environment variables.

Then enable verbose mode ( setting msglvl == 1 ) and see the perf results for each of phases ( reordering, factorization, solving) and see the scalability results if any or any performance bottleneck

--Gennady



0 Kudos
Copy link
Reply

Community support is provided during standard business hours (Monday to Friday 7AM - 5PM PST). 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.