https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/OMP-threading/m-p/975589#M17054 <DIV>Hi,</DIV> <DIV></DIV> <DIV> We are using only a selected few routines from Intel MKL such as dgemm, daxpy, ddot, dgemv, etc. For all the other blas/lapack routines we compile and link in the actual lapack/blas routines. My questions are as follows:</DIV> <DIV></DIV> <DIV> (i) Is dgemm, daxpy, ddot or dgemv parallelized ?</DIV> <DIV> (ii) Do I need to use the full suite of Intel MKL routines to take advantage of parallelism in a select few routines?</DIV> <DIV></DIV> <DIV>By the way, we are using Intel MKL 7.2</DIV> <DIV></DIV> <DIV>Easwaran</DIV> Wed, 14 Sep 2005 21:43:19 GMT easwarv 2005-09-14T21:43:19Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/OMP-threading/m-p/975589#M17054 <DIV>Hi,</DIV> <DIV></DIV> <DIV> We are using only a selected few routines from Intel MKL such as dgemm, daxpy, ddot, dgemv, etc. For all the other blas/lapack routines we compile and link in the actual lapack/blas routines. My questions are as follows:</DIV> <DIV></DIV> <DIV> (i) Is dgemm, daxpy, ddot or dgemv parallelized ?</DIV> <DIV> (ii) Do I need to use the full suite of Intel MKL routines to take advantage of parallelism in a select few routines?</DIV> <DIV></DIV> <DIV>By the way, we are using Intel MKL 7.2</DIV> <DIV></DIV> <DIV>Easwaran</DIV> Wed, 14 Sep 2005 21:43:19 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/OMP-threading/m-p/975589#M17054 easwarv 2005-09-14T21:43:19Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/OMP-threading/m-p/975590#M17055 <DIV>Why use MKL for some functions, but not for others? I'm curious.</DIV> <DIV></DIV> <DIV>DGEMM is certainly parallized. The other functions from Levels 1 and 2 of the BLAS are not parallelized.</DIV> <DIV></DIV> <DIV>-Todd</DIV> Wed, 14 Sep 2005 23:08:47 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/OMP-threading/m-p/975590#M17055 Todd_R_Intel 2005-09-14T23:08:47Z