https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fat-and-Narrow-matrix-multiplication-with-quot-cblas-cgemm-quot/m-p/1171705#M28622 <P>What is it that you want us to disagree with? It is quite plausible that there exists/existed some computer and some version of MKL with&nbsp;which the task that you stated took 30 ms to execute. Why is that worth discussing?</P> Tue, 28 Apr 2020 11:51:41 GMT mecej4 2020-04-28T11:51:41Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fat-and-Narrow-matrix-multiplication-with-quot-cblas-cgemm-quot/m-p/1171704#M28621 <P>Hi,&nbsp;</P><P>I am working with a matrix multiplication of sizes A = 40 x 40 and B is 40 x 10k with MKL support functions "<STRONG>cblas_cgemm</STRONG>". It is taking a 30 milliseconds,</P><P>I have enabled mkl multithreading also, which I belive it is more.</P><P>I have read in internet that "MKL functions are optimized for generic matrix multiplications"..&nbsp;</P><P>Anybody agrees or disagrees with me.</P><P>&nbsp;</P><P>Thanks in advance .</P> Sun, 26 Apr 2020 06:20:42 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fat-and-Narrow-matrix-multiplication-with-quot-cblas-cgemm-quot/m-p/1171704#M28621 sivasankar__mamilla 2020-04-26T06:20:42Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fat-and-Narrow-matrix-multiplication-with-quot-cblas-cgemm-quot/m-p/1171705#M28622 <P>What is it that you want us to disagree with? It is quite plausible that there exists/existed some computer and some version of MKL with&nbsp;which the task that you stated took 30 ms to execute. Why is that worth discussing?</P> Tue, 28 Apr 2020 11:51:41 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Fat-and-Narrow-matrix-multiplication-with-quot-cblas-cgemm-quot/m-p/1171705#M28622 mecej4 2020-04-28T11:51:41Z