<?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: Change number of threads for pardiso in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862160#M7582</link>
    <description>Dong,&lt;BR /&gt;&lt;BR /&gt;&amp;gt;The draw back of the above method is that I have to re-do the factoriation. Is there a better way to do it?&lt;BR /&gt;&lt;BR /&gt;No - this functionality is very useful but it is not yet implemented in the current version of PARDISO.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;Olaf &lt;BR /&gt;</description>
    <pubDate>Mon, 07 Jul 2008 10:41:04 GMT</pubDate>
    <dc:creator>basel</dc:creator>
    <dc:date>2008-07-07T10:41:04Z</dc:date>
    <item>
      <title>Change number of threads for pardiso</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862159#M7581</link>
      <description>&lt;P&gt;I'm using MKL10.0.1.014 and testing MKL 10.1.0.006 on Linux. It seems that when evern I use "omp_set_num_threads" and/or "mkl_set_num_threads" to change the number of threads, MKL pardiso crashes.&lt;/P&gt;
&lt;P&gt;The only way I can make it work is by doing the following&lt;/P&gt;
&lt;P&gt;Use "mkl_set_num_threads" to change the number of threads&lt;BR /&gt;Unload pardiso with "phase=-1"&lt;BR /&gt;Reinitiate pardiso with "phase=11"&lt;BR /&gt;Re-do factoriation.&lt;/P&gt;
&lt;P&gt;The draw back of the above method is that I have to re-do the factoriation. Is there a better way to do it?&lt;/P&gt;
&lt;P&gt;DZ&lt;/P&gt;</description>
      <pubDate>Mon, 07 Jul 2008 09:15:15 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862159#M7581</guid>
      <dc:creator>Dong_Zheng</dc:creator>
      <dc:date>2008-07-07T09:15:15Z</dc:date>
    </item>
    <item>
      <title>Re: Change number of threads for pardiso</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862160#M7582</link>
      <description>Dong,&lt;BR /&gt;&lt;BR /&gt;&amp;gt;The draw back of the above method is that I have to re-do the factoriation. Is there a better way to do it?&lt;BR /&gt;&lt;BR /&gt;No - this functionality is very useful but it is not yet implemented in the current version of PARDISO.&lt;BR /&gt;&lt;BR /&gt;Regards,&lt;BR /&gt;Olaf &lt;BR /&gt;</description>
      <pubDate>Mon, 07 Jul 2008 10:41:04 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862160#M7582</guid>
      <dc:creator>basel</dc:creator>
      <dc:date>2008-07-07T10:41:04Z</dc:date>
    </item>
    <item>
      <title>Re: Change number of threads for pardiso</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862161#M7583</link>
      <description>&lt;P&gt;Dong &lt;/P&gt;
&lt;P&gt;Would you please provide more details about the failure? In particular I'd like to take a look at iparm settings. &lt;/P&gt;
&lt;P&gt;It looks like you are trying to use out-of-core version with iparm(60)=1. For this particular case mkl_num_threads has to be 1 since our out-of-core version has not been parallelized yet. &lt;/P&gt;
&lt;P&gt;Thanks in advance&lt;/P&gt;
&lt;P&gt;All the best&lt;/P&gt;
&lt;P&gt;Sergey&lt;/P&gt;</description>
      <pubDate>Tue, 08 Jul 2008 10:15:47 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Change-number-of-threads-for-pardiso/m-p/862161#M7583</guid>
      <dc:creator>Sergey_K_Intel1</dc:creator>
      <dc:date>2008-07-08T10:15:47Z</dc:date>
    </item>
  </channel>
</rss>

