topic Thank you, Gennady in Intel® oneAPI Math Kernel Library

MKL sparse BLAS

Roberto_Pinto_Souto — Mon, 08 Jun 2015 19:37:39 GMT

Hi all,

Please, are threaded the following MKL sparse BLAS functions?
'mkl_cspblas_?coosymv', 'mkl_?coosymv' and 'mkl_sparse_?_mv1 (11.3 beta)

I'm not able to run with more than one thread these functions.

Thanks in advance,

Roberto Souto.

what do you mean by mkl

Gennady_F_Intel — Tue, 09 Jun 2015 06:27:02 GMT

what do you mean by mkl_sparse_?_mv1?

Dear Gennady Fedorov.

Roberto_Pinto_Souto — Tue, 09 Jun 2015 08:32:19 GMT

Dear Gennady Fedorov.

I'm sorry. I typed wrong. The function is mkl_sparse_?_mv. More specifically mkl_sparse_d_mv, from Intel® Math Kernel Library Inspector-executor Sparse BLAS Routines (https://software.intel.com/en-us/articles/intel-math-kernel-library-inspector-executor-sparse-blas-routines)

I'm not able to run this function more than one thread, even for OMP_NUM_THREADS set greater than 1. The same occur with the functions mkl_cspblas_dcoosymv and mkl_dcoosymv;

Thank you,

Roberto.

Actually all of these

Gennady_F_Intel — Tue, 09 Jun 2015 10:24:58 GMT

Actually all of these functions are not threaded internally at this moment ( including MKL v,113. beta ).

Thank you, Gennady

Roberto_Pinto_Souto — Tue, 09 Jun 2015 18:28:24 GMT

Thank you, Gennady

Best regards,

Roberto Souto