<?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 pardiso crashed with no memory when only processing 1 million * 1 million sparse matrix in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/pardiso-crashed-with-no-memory-when-only-processing-1-million-1/m-p/907603#M11858</link>
    <description>Hi, &lt;BR /&gt;&lt;BR /&gt;I use pardiso to solve a million cells problem, i.e. the matrix size is only 1 million * 1million, the nozero number is only 3.9 million numbers, in my pc with 2G RAM. &lt;BR /&gt;&lt;BR /&gt;Unfortunately, pardiso crashes during LU factoration. Pardiso tells me during reordering in partition metis it produces nearly 800 million nozeros in L+U. &lt;BR /&gt;&lt;BR /&gt;It seems to be a little ridiculous, right? In LU0 factoration, I think the nozero number should not exceed the original matrix's nozero. &lt;BR /&gt;&lt;BR /&gt;I do think some parameters are set to be wrong. But I do not know which parameter should be responsible for such case , can anyone point me out for this problem?&lt;BR /&gt;&lt;BR /&gt;Thanks very much&lt;BR /&gt;</description>
    <pubDate>Wed, 06 Aug 2008 06:44:39 GMT</pubDate>
    <dc:creator>smth</dc:creator>
    <dc:date>2008-08-06T06:44:39Z</dc:date>
    <item>
      <title>pardiso crashed with no memory when only processing 1 million * 1 million sparse matrix</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/pardiso-crashed-with-no-memory-when-only-processing-1-million-1/m-p/907603#M11858</link>
      <description>Hi, &lt;BR /&gt;&lt;BR /&gt;I use pardiso to solve a million cells problem, i.e. the matrix size is only 1 million * 1million, the nozero number is only 3.9 million numbers, in my pc with 2G RAM. &lt;BR /&gt;&lt;BR /&gt;Unfortunately, pardiso crashes during LU factoration. Pardiso tells me during reordering in partition metis it produces nearly 800 million nozeros in L+U. &lt;BR /&gt;&lt;BR /&gt;It seems to be a little ridiculous, right? In LU0 factoration, I think the nozero number should not exceed the original matrix's nozero. &lt;BR /&gt;&lt;BR /&gt;I do think some parameters are set to be wrong. But I do not know which parameter should be responsible for such case , can anyone point me out for this problem?&lt;BR /&gt;&lt;BR /&gt;Thanks very much&lt;BR /&gt;</description>
      <pubDate>Wed, 06 Aug 2008 06:44:39 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/pardiso-crashed-with-no-memory-when-only-processing-1-million-1/m-p/907603#M11858</guid>
      <dc:creator>smth</dc:creator>
      <dc:date>2008-08-06T06:44:39Z</dc:date>
    </item>
    <item>
      <title>Re: pardiso crashed with no memory when only processing 1 milli</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/pardiso-crashed-with-no-memory-when-only-processing-1-million-1/m-p/907604#M11859</link>
      <description>&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;Hello,&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;What MKL version are you using?&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;Since version 10.0, MKL support out-of-core version (OOC).&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;The OOC &lt;SPAN&gt;PARDISO&lt;/SPAN&gt; can solve very large problems by holding the matrix factors in files on the disk.&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;So, please set the parameters as: &lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN class="parmname"&gt;&lt;SPAN&gt;iparm&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;SAMP&gt;&lt;SPAN&gt;(3)&lt;/SPAN&gt;&lt;/SAMP&gt;&lt;SPAN&gt; = 1&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN class="parmname"&gt;&lt;SPAN&gt;iparm&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;SAMP&gt;&lt;SPAN&gt;(60)= 2&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/SAMP&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;For Fortran&lt;/SPAN&gt;&lt;SAMP&gt;&lt;SPAN&gt;Or&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/SAMP&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SAMP&gt;&lt;SPAN&gt;&lt;SPAN&gt;&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/SAMP&gt;&lt;SPAN class="parmname"&gt;&lt;SPAN&gt;Iparm[2]&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;SPAN&gt; = 1&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN class="parmname"&gt;&lt;SPAN&gt;Iparm[59] =&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;SAMP&gt;&lt;SPAN&gt;2&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/SAMP&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SAMP&gt;&lt;SPAN&gt;for c/c++:&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/SAMP&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;&lt;P&gt;&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;&lt;P&gt;and try to use pardiso ooc. it should work. :)&lt;/P&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P class="MsoNormal"&gt;&lt;SPAN&gt;regards, Gennady&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Thu, 28 Aug 2008 12:34:48 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/pardiso-crashed-with-no-memory-when-only-processing-1-million-1/m-p/907604#M11859</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2008-08-28T12:34:48Z</dc:date>
    </item>
  </channel>
</rss>

