<?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 in OOC mode uses same memory as in IC mode in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279872#M31293</link>
    <description>&lt;P&gt;Hi Kirill, thank you for the quick response.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Swap size size does not have an effect.&amp;nbsp; However, setting iparm(24)=1 (I am using fortran) does reduce the memory consumption in OOC mode.&amp;nbsp; I will run larger matrices over the next few days and report behavior.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;iparm(24)=1 causes segfault whith pardiso_handle_store, which I would also like to use.&amp;nbsp; Is there a way to use iparm(24)=1 with pardiso_handle_store?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Attached is a sample code and matrix data to reproduce the results.&amp;nbsp; I had to use a much smaller matrix than the original post due to the upload size limits.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
    <pubDate>Fri, 07 May 2021 20:00:15 GMT</pubDate>
    <dc:creator>Pan11</dc:creator>
    <dc:date>2021-05-07T20:00:15Z</dc:date>
    <item>
      <title>Pardiso in OOC mode uses same memory as in IC mode</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279776#M31289</link>
      <description>&lt;P&gt;I am trying to use pardiso in ooc mode with iparm(60)=2 from fortran.&amp;nbsp; It generates the ooc files but consumes the same ram as in ic mode iparm(60)=0.&amp;nbsp; Behavior is the same regardless of the size of the matrix, total or available mem, number or threads, or computer. Tested in centos 7 and 8 using 20.2 relase. Attached is the msglvl=1 output from runs for the same matrix.&amp;nbsp; Available mem scanned from /proc/meminfo is listed as well.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thanks for any help.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Fri, 07 May 2021 12:45:22 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279776#M31289</guid>
      <dc:creator>Pan11</dc:creator>
      <dc:date>2021-05-07T12:45:22Z</dc:date>
    </item>
    <item>
      <title>Re: Pardiso in OOC mode uses same memory as in IC mode</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279829#M31291</link>
      <description>&lt;P&gt;Hello!&lt;/P&gt;
&lt;P&gt;I don't see anything bad about the output you shared, which would explain what you see. One thing maybe: try to leave the swap size set to 0 rather than setting it to something else.&lt;/P&gt;
&lt;P&gt;As a quick attempt, can you also try with iparm[23]=1 (so that you don't get the single-level factorization algorithm but rather a two-level one) and let us know the results?&lt;/P&gt;
&lt;P&gt;Independent from whether this helps or not, I think we need a reproducer so that we can check what happens on our side.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Best,&lt;BR /&gt;Kirill&lt;/P&gt;</description>
      <pubDate>Fri, 07 May 2021 17:14:53 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279829#M31291</guid>
      <dc:creator>Kirill_V_Intel</dc:creator>
      <dc:date>2021-05-07T17:14:53Z</dc:date>
    </item>
    <item>
      <title>Re: Pardiso in OOC mode uses same memory as in IC mode</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279872#M31293</link>
      <description>&lt;P&gt;Hi Kirill, thank you for the quick response.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Swap size size does not have an effect.&amp;nbsp; However, setting iparm(24)=1 (I am using fortran) does reduce the memory consumption in OOC mode.&amp;nbsp; I will run larger matrices over the next few days and report behavior.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;iparm(24)=1 causes segfault whith pardiso_handle_store, which I would also like to use.&amp;nbsp; Is there a way to use iparm(24)=1 with pardiso_handle_store?&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Attached is a sample code and matrix data to reproduce the results.&amp;nbsp; I had to use a much smaller matrix than the original post due to the upload size limits.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;</description>
      <pubDate>Fri, 07 May 2021 20:00:15 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279872#M31293</guid>
      <dc:creator>Pan11</dc:creator>
      <dc:date>2021-05-07T20:00:15Z</dc:date>
    </item>
    <item>
      <title>Re:Pardiso in OOC mode uses same memory as in IC mode</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279938#M31296</link>
      <description>&lt;P&gt;regarding the uploading size limit -- we have an official support channel - Intel Online Service Center (  &lt;A href="https://supporttickets.intel.com/servicecenter?lang=en-US" target="_blank"&gt;https://supporttickets.intel.com/servicecenter?lang=en-US&lt;/A&gt;) and You can open the ticket there against MKL. The uploading size limitation there &amp;lt;= 2Gb.&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Sat, 08 May 2021 03:20:24 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1279938#M31296</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2021-05-08T03:20:24Z</dc:date>
    </item>
    <item>
      <title>Re: Re:Pardiso in OOC mode uses same memory as in IC mode</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1280278#M31303</link>
      <description>&lt;P&gt;Hi Gennady, thanks for the info.&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Also, I have run several large tests and confirmed OOC mode with iparm(24)=1 consumes the memory estimated by iparm(63) mem in phase 22.&amp;nbsp; I will start another thread for the pardiso_handle_store issues.&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thank you!&lt;/P&gt;</description>
      <pubDate>Mon, 10 May 2021 17:37:06 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1280278#M31303</guid>
      <dc:creator>Pan11</dc:creator>
      <dc:date>2021-05-10T17:37:06Z</dc:date>
    </item>
    <item>
      <title>Re:Pardiso in OOC mode uses same memory as in IC mode</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1280468#M31305</link>
      <description>&lt;P&gt;The issue is closing and we will no longer respond to this thread.&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Tue, 11 May 2021 07:40:35 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Pardiso-in-OOC-mode-uses-same-memory-as-in-IC-mode/m-p/1280468#M31305</guid>
      <dc:creator>Gennady_F_Intel</dc:creator>
      <dc:date>2021-05-11T07:40:35Z</dc:date>
    </item>
  </channel>
</rss>

