https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864234#M7734 <DIV style="margin:0px;"></DIV> <BR />Hello,<BR /><BR />Unfortunately, you cannot use PARDISO in a such way. It needs explicitly all matrix elements in order to make LLT or LU decompositions.<BR /><BR />Artur gave yougood advice: use iterative solvers which support reverse communication interface.<BR /><BR />Regards,<BR />Konstantin Tue, 06 Oct 2009 07:09:34 GMT Konstantin_A_Intel 2009-10-06T07:09:34Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864232#M7732 I want to use the PARDISO.<BR /><BR />However, <STRONG>the matrix to be solved does not be explicitlygenerated</STRONG>, for example, the fast multiple method(FMM). In other words,I does not know the entries of the matrix,although<STRONG> I can carry out the matrix-vector multiplication</STRONG>.<BR /><BR />So, the matrix itself can not be a input parameter for PARDISO.<BR /><BR /><STRONG>Can I perform matrix-vector multiplication myself when using PARDISO ?</STRONG> Wed, 23 Sep 2009 14:37:54 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864232#M7732 afd_lml 2009-09-23T14:37:54Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864233#M7733 <DIV style="margin:0px;"> <DIV id="quote_reply" style="width: 100%; margin-top: 5px;"> <DIV style="margin-left:2px;margin-right:2px;">Quoting - <A href="https://community.intel.com/en-us/profile/441618">afd.lml</A></DIV> <DIV style="background-color:#E5E5E5; padding:5px;border: 1px; border-style: inset;margin-left:2px;margin-right:2px;"><EM>I want to use the PARDISO.<BR /><BR />However, <STRONG>the matrix to be solved does not be explicitlygenerated</STRONG>, for example, the fast multiple method(FMM). In other words,I does not know the entries of the matrix,although<STRONG> I can carry out the matrix-vector multiplication</STRONG>.<BR /><BR />So, the matrix itself can not be a input parameter for PARDISO.<BR /><BR /><STRONG>Can I perform matrix-vector multiplication myself when using PARDISO ?</STRONG> </EM></DIV> </DIV> </DIV> Usually FMM employs GMRES based algorithm for solution procedure. MKL offers FGMRES which allows you (actually requires, it is reverse communication type) to perform matrix-vector multiplication yourself.<BR /><BR />I'm unaware of any way that PARDISO can be used in that fashion. Presumably somebody will provide some more definite answer to that.<BR /><BR />A.<BR /> Thu, 24 Sep 2009 00:33:02 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864233#M7733 ArturGuzik 2009-09-24T00:33:02Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864234#M7734 <DIV style="margin:0px;"></DIV> <BR />Hello,<BR /><BR />Unfortunately, you cannot use PARDISO in a such way. It needs explicitly all matrix elements in order to make LLT or LU decompositions.<BR /><BR />Artur gave yougood advice: use iterative solvers which support reverse communication interface.<BR /><BR />Regards,<BR />Konstantin Tue, 06 Oct 2009 07:09:34 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Can-I-perform-matrix-vector-multiplication-myself-when-using/m-p/864234#M7734 Konstantin_A_Intel 2009-10-06T07:09:34Z