https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/eigenvalue-problem-and-threading/m-p/902300#M11345 <PRE>I am using LAPACK routines in order to determine eigenvalues<BR />of symmetric or hermitean matrices. These routines are contained<BR />in the MKL and in principle allow threads. Looking at the<BR />perfomance I realize that they are not or only poorly parallelized.<BR /><BR />Does anybody know whether one of the routines<BR />ZHEEV, ZHEEVD, or ZHEEVR or a similar one<BR />has been worked over to use multiple threads for<BR />matrix diagonalization?<BR /><BR />Thanks a lot, Juergen Schnack (Osnabrueck, Germany)<BR /><BR /><BR /></PRE> Sat, 04 Nov 2006 00:19:18 GMT jschnack 2006-11-04T00:19:18Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/eigenvalue-problem-and-threading/m-p/902300#M11345 <PRE>I am using LAPACK routines in order to determine eigenvalues<BR />of symmetric or hermitean matrices. These routines are contained<BR />in the MKL and in principle allow threads. Looking at the<BR />perfomance I realize that they are not or only poorly parallelized.<BR /><BR />Does anybody know whether one of the routines<BR />ZHEEV, ZHEEVD, or ZHEEVR or a similar one<BR />has been worked over to use multiple threads for<BR />matrix diagonalization?<BR /><BR />Thanks a lot, Juergen Schnack (Osnabrueck, Germany)<BR /><BR /><BR /></PRE> Sat, 04 Nov 2006 00:19:18 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/eigenvalue-problem-and-threading/m-p/902300#M11345 jschnack 2006-11-04T00:19:18Z