<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" version="2.0">
  <channel>
    <title>topic Re: PARDISO for Finite Element in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868209#M8182</link>
    <description>&lt;DIV style="margin:0px;"&gt;
&lt;DIV id="quote_reply" style="width: 100%; margin-top: 5px;"&gt;
&lt;DIV style="margin-left:2px;margin-right:2px;"&gt;Quoting - &lt;A href="https://community.intel.com/en-us/profile/411664"&gt;jjforest&lt;/A&gt;&lt;/DIV&gt;
&lt;DIV style="background-color:#E5E5E5; padding:5px;border: 1px; border-style: inset;margin-left:2px;margin-right:2px;"&gt;&lt;EM&gt; Thank you Gennady, mkl_progress() does the trick of reporting factorization progress. I have not bought the product and Premier support yet, so I can not submit my suggestions as a feature request. I will try to remember that after I purchase the product.&lt;BR /&gt;&lt;/EM&gt;&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;BR /&gt;My program freezes at SetWindowText() in the following code on Vista (not XP) machine. Any ideas?&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;extern "C" int MKL_PROGRESS( int* ithr, int* step, char* stage, int len )&lt;BR /&gt;{  // this is callback function from FORTRAN PARDISO code&lt;BR /&gt;&lt;BR /&gt; if(gpPercentWnd)&lt;BR /&gt; {&lt;BR /&gt;&lt;BR /&gt; OutputDebugStringA("!!MKL_1");&lt;BR /&gt; CString strMsg;&lt;BR /&gt; strMsg.Format(_T("Progress %d%%"), *step);&lt;BR /&gt; gpPercentWnd-&amp;gt;SetWindowText(strMsg);&lt;BR /&gt; OutputDebugStringA("!!MKL_2");&lt;BR /&gt; }&lt;BR /&gt; return 0;&lt;BR /&gt;}&lt;BR /&gt;</description>
    <pubDate>Thu, 12 Mar 2009 14:49:58 GMT</pubDate>
    <dc:creator>jjforest</dc:creator>
    <dc:date>2009-03-12T14:49:58Z</dc:date>
    <item>
      <title>PARDISO for Finite Element</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868206#M8179</link>
      <description>I am going to incorporate PARDISO in our finite element package. I am very happy with PARDISCO's efficiency using the evaluation version. I do have the following suggestions.&lt;BR /&gt;&lt;BR /&gt;1. PARDISO should provide more extensive error messages. In particular, when the matrix is ill-conditioned, the solver should return which diagonal element is in trouble (not just return error = -4). This is important because finite element package users often make modeling errors that result in ill-conditioned stiffness matrix. The users need to be informed where (which nodes) the problem lies.&lt;BR /&gt;&lt;BR /&gt;2. PARDISO should provide quad (double-double) precision in addition to the double precision math. In some finite element problems we encounter, double precision will result wrong solution. I can provide you a sample model of such case if you want.&lt;BR /&gt;&lt;BR /&gt;3. It would be nice if PARDISO take a callback function that tells the current progress (1%, 2%, etc) of factorization. I am not sure if this is possible with FORTRAN since I am not very familar with FORTRAN language. I program in C/C++.&lt;BR /&gt;&lt;BR /&gt;Thanks for your consideration</description>
      <pubDate>Mon, 12 Jan 2009 23:52:41 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868206#M8179</guid>
      <dc:creator>jjforest</dc:creator>
      <dc:date>2009-01-12T23:52:41Z</dc:date>
    </item>
    <item>
      <title>Re: PARDISO for Finite Element</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868207#M8180</link>
      <description>&lt;BR /&gt;
&lt;P&gt;Hi,&lt;/P&gt;
&lt;OL type="1"&gt;
&lt;LI&gt;Your suggestions make sense and I would recommend you submit your suggestions as a Feature Requests against MKL to Premier support( &lt;A title="https://premier.intel.com/" href="https://premier.intel.com/"&gt;https://premier.intel.com/&lt;/A&gt; ) &lt;/LI&gt;
&lt;/OL&gt;
&lt;P&gt; "PARDISO should provide more extensive error messages ...." and&lt;/P&gt;
&lt;P&gt; "PARDISO should provide quad precision" and you can add the sample model of this case.&lt;/P&gt;
&lt;OL type="1"&gt;
&lt;LI&gt;this functionality is available already in MKL version.10.1: 
&lt;UL type="disc"&gt;
&lt;LI&gt;see PARDISO's input parameter description: msglvl - Message Level Information&lt;/LI&gt;
&lt;/UL&gt;
&lt;/LI&gt;
&lt;/OL&gt;
&lt;P&gt; if msglvl == 0, then PARDISO generates no output,&lt;/P&gt;
&lt;P&gt; if msglvl == 1, the solver prints statistical information to the screen&lt;/P&gt;
&lt;P&gt; or you can use mkl_progress() routine for that.&lt;/P&gt;
--Gennady</description>
      <pubDate>Tue, 13 Jan 2009 17:58:36 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868207#M8180</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2009-01-13T17:58:36Z</dc:date>
    </item>
    <item>
      <title>Re: PARDISO for Finite Element</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868208#M8181</link>
      <description>&lt;DIV style="margin:0px;"&gt;&lt;/DIV&gt;
Thank you Gennady, mkl_progress() does the trick of reporting factorization progress. I have not bought the product and Premier support yet, so I can not submit my suggestions as a feature request. I will try to remember that after I purchase the product.&lt;BR /&gt;</description>
      <pubDate>Wed, 14 Jan 2009 01:30:44 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868208#M8181</guid>
      <dc:creator>jjforest</dc:creator>
      <dc:date>2009-01-14T01:30:44Z</dc:date>
    </item>
    <item>
      <title>Re: PARDISO for Finite Element</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868209#M8182</link>
      <description>&lt;DIV style="margin:0px;"&gt;
&lt;DIV id="quote_reply" style="width: 100%; margin-top: 5px;"&gt;
&lt;DIV style="margin-left:2px;margin-right:2px;"&gt;Quoting - &lt;A href="https://community.intel.com/en-us/profile/411664"&gt;jjforest&lt;/A&gt;&lt;/DIV&gt;
&lt;DIV style="background-color:#E5E5E5; padding:5px;border: 1px; border-style: inset;margin-left:2px;margin-right:2px;"&gt;&lt;EM&gt; Thank you Gennady, mkl_progress() does the trick of reporting factorization progress. I have not bought the product and Premier support yet, so I can not submit my suggestions as a feature request. I will try to remember that after I purchase the product.&lt;BR /&gt;&lt;/EM&gt;&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;/DIV&gt;
&lt;BR /&gt;My program freezes at SetWindowText() in the following code on Vista (not XP) machine. Any ideas?&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;extern "C" int MKL_PROGRESS( int* ithr, int* step, char* stage, int len )&lt;BR /&gt;{  // this is callback function from FORTRAN PARDISO code&lt;BR /&gt;&lt;BR /&gt; if(gpPercentWnd)&lt;BR /&gt; {&lt;BR /&gt;&lt;BR /&gt; OutputDebugStringA("!!MKL_1");&lt;BR /&gt; CString strMsg;&lt;BR /&gt; strMsg.Format(_T("Progress %d%%"), *step);&lt;BR /&gt; gpPercentWnd-&amp;gt;SetWindowText(strMsg);&lt;BR /&gt; OutputDebugStringA("!!MKL_2");&lt;BR /&gt; }&lt;BR /&gt; return 0;&lt;BR /&gt;}&lt;BR /&gt;</description>
      <pubDate>Thu, 12 Mar 2009 14:49:58 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/PARDISO-for-Finite-Element/m-p/868209#M8182</guid>
      <dc:creator>jjforest</dc:creator>
      <dc:date>2009-03-12T14:49:58Z</dc:date>
    </item>
  </channel>
</rss>

