Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
6981 Discussions

about hybrid multithreaded-MPI solver

recrusader
Beginner
‎08-10-2010 07:47 AM
341 Views
I am wondering whether it is possible to use MKL and Intel MPI to realize hybrid multithreaded-MPI solver with simple codes the users write. Thanks a lot.
0 Kudos

Link Copied

2 Replies
TimP
Honored Contributor III
‎08-10-2010 03:21 PM
341 Views
Yes, there are several production codes as well as demonstrations of Intel MPI usage with Intel OpenMP as the MPI FUNNELED implementation. Currently supported versions of Intel MPI default to using the OMP_NUM_THREADS to optimize thread placement on each node. MKL works well in both the 1 MPI process per node and 1 process per socket modes.
0 Kudos
Copy link
Dmitry_K_Intel2
Employee
‎08-11-2010 12:23 AM
341 Views
Intel MKL from version 10.2 and Intel MPI Library from version 4.0 may balance workload on a node. Number of MPI processes plus number of MKL threads will not exceed number of cores on a node. You don't even need to use OMP_NUM_THREADS or MKL_NUM_THREADS.

Give it a try!

Regards!
Dmitry
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.