https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Speed-of-DGGHRD-and-DGGHD3/m-p/1197716#M29854 <P>Update, it looks like the blocksize returned by ilaenv is always 1 for DGGHD3. Changing that seems to fix the performance issue.</P> Wed, 05 Aug 2020 17:32:08 GMT thijssteel 2020-08-05T17:32:08Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Speed-of-DGGHRD-and-DGGHD3/m-p/1197564#M29852 <P>I noticed something about the routines DGGHRD and DGGHD3. They both reduce a pencil to Hessenberg triangular form for use in the QZ algorithm, with DGGHD3 being the blocked version.</P> <P>When linked with the reference LAPACK or with OpenBLAS, DGGHD3 is faster than DHHGRD. However, when the test program is linked with MKL, DGGHD3 loses its speed advantage. I also don't see any multicore usage in that routine when linked with MKL. Is it possible that the blocked version is not (fully) implemented in MKL?</P> Wed, 05 Aug 2020 09:11:36 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Speed-of-DGGHRD-and-DGGHD3/m-p/1197564#M29852 thijssteel 2020-08-05T09:11:36Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Speed-of-DGGHRD-and-DGGHD3/m-p/1197716#M29854 <P>Update, it looks like the blocksize returned by ilaenv is always 1 for DGGHD3. Changing that seems to fix the performance issue.</P> Wed, 05 Aug 2020 17:32:08 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Speed-of-DGGHRD-and-DGGHD3/m-p/1197716#M29854 thijssteel 2020-08-05T17:32:08Z