ZYTRI sequential?

Olaf_Schenk — Fri, 17 Jul 2015 14:26:50 GMT

Hi
I am testing the following code with composer_xe_2015

lwork = 256*n
call ZSYTRF( 'U', n, afull, n, ipiv, work, lwork, error )
call ZSYTRI( 'U', n, afull, n, ipiv, work, lwork, error )

using

gfortran -O3 -fopenmp read_blas.f90 -Wl,--start-group ${MKLROOT}/lib/intel64/libmkl_gf_lp64.a ${MKLROOT}/lib/intel64/libmkl_core.a ${MKLROOT}/lib/intel64/libmkl_gnu_thread.a -Wl,--end-group -ldl -lpthread -lm

After setting OMP_NUM_THREADS=16 I can see in "top" that ZSYTRF runs in parallel but ZSYTRI not. The matrix is large enough (n=15120)

Any idea how to solve this problem?

Olaf

Hi Olaf

Ying_H_Intel — Mon, 20 Jul 2015 03:00:55 GMT

Hi Olaf

You are right, the function is not threaded. Here is threaded function list for your reference.

https://software.intel.com/en-us/articles/intel-mkl-threaded-functions/

As i understand, you actually expect the function is parallel so improve the performance, right? If yes, could you please let me know some information, like background, problem size, hardware etc. If private information, you may reply me by Private message.

Thanks

Ying

topic Hi Olaf in Intel® oneAPI Math Kernel Library

ZYTRI sequential?

Hi Olaf