<?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 Large Scale Problem using PARDISO. in Intel® oneAPI Math Kernel Library</title>
    <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779076#M1286</link>
    <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I am solving &lt;A&gt;&lt;X&gt; = &lt;B&gt; where both X and B are matrices. I 
am solving for X. Size of &lt;/B&gt;&lt;/X&gt;&lt;/A&gt;&lt;A&gt; is 1,500,000 x 1,500,000. Aproximate non 
zeros are 60,000,000. I am able to run my code for the lower size 
problem. But for this size it does not run.&lt;BR /&gt;&lt;BR /&gt;System configuration:&lt;BR /&gt;&lt;BR /&gt;SGI - 8 processor with 32 GB memory.&lt;BR /&gt;Compiler - GNU.&lt;BR /&gt;&lt;BR /&gt;I am using 10.0.1.014 intel mkl&lt;BR /&gt;&lt;BR /&gt;When I am using iparm(60) = 2 it kill the process with -9 error&lt;BR /&gt;&lt;BR /&gt;***************************FOLLOWING WAS ERROR******************************&lt;BR /&gt;&lt;BR /&gt;The file ./pardiso_ooc.cfg was not opened&lt;BR /&gt;*** error PARDISO ( not enough mem. for OOC) error_num= -9&lt;BR /&gt;&lt;BR /&gt;================ PARDISO: solving a symmetric indef. system ================&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Summary PARDISO: ( reorder to reorder )&lt;BR /&gt;================&lt;BR /&gt;&lt;BR /&gt;Times:&lt;BR /&gt;======&lt;BR /&gt;&lt;BR /&gt; Time fulladj: 1.360909 s&lt;BR /&gt; Time reorder: 29.476320 s&lt;BR /&gt; Time symbfct: 34.058441 s&lt;BR /&gt; Time malloc : 0.000549 s&lt;BR /&gt; Time total : 64.896383 s total - sum: 0.000164 s&lt;BR /&gt;&lt;BR /&gt;Statistics:&lt;BR /&gt;===========&lt;BR /&gt;&amp;lt; Parallel Direct Factorization with #processors: &amp;gt; 1&lt;BR /&gt;&amp;lt; Numerical Factorization with Level-3 BLAS performance &amp;gt;&lt;BR /&gt;&lt;BR /&gt;&amp;lt; Linear system Ax = b&amp;gt; &lt;TRANSPOSE&gt;&lt;BR /&gt; #equations: 1460346&lt;BR /&gt; #non-zeros in A: 58022798&lt;BR /&gt; non-zeros in A (%): 0.002721&lt;BR /&gt; #right-hand sides: 1&lt;BR /&gt;&lt;BR /&gt;&amp;lt; Factors L and U &amp;gt;&lt;BR /&gt; #columns for each panel: 10&lt;BR /&gt; #independent subgraphs: 0&lt;BR /&gt;&amp;lt; Preprocessing with state of the art partitioning metis&amp;gt;&lt;BR /&gt; #supernodes: 231589&lt;BR /&gt; size of largest supernode: 20746&lt;BR /&gt; number of nonzeros in L -905604445&lt;BR /&gt; number of nonzeros in U 1&lt;BR /&gt; number of nonzeros in L+U -905604444&lt;BR /&gt;test&lt;BR /&gt;check point 1&lt;BR /&gt;check point 2&lt;BR /&gt;check point 3&lt;BR /&gt;check point 5&lt;BR /&gt;check point 7&lt;BR /&gt; 58022799&lt;BR /&gt;Reordering completed ... &lt;BR /&gt;The following ERROR was detected: -9&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;*****************************************************************************&lt;BR /&gt;Questions:&lt;BR /&gt;&lt;BR /&gt;1) Is it possible to solve using current resources (8-Processors with each of 4GB RAM)?&lt;BR /&gt;2)
 If I want to run this size problem then what is the maximum RAM on 
single processor will I need? As we are planning to buy a system that 
has sufficient memory for this type of operations.&lt;BR /&gt;&lt;BR /&gt;Thank you,&lt;BR /&gt;&lt;BR /&gt;Mital Mistry.&lt;/TRANSPOSE&gt;&lt;/A&gt;</description>
    <pubDate>Mon, 01 Aug 2011 14:13:20 GMT</pubDate>
    <dc:creator>Mistry__Mital</dc:creator>
    <dc:date>2011-08-01T14:13:20Z</dc:date>
    <item>
      <title>Large Scale Problem using PARDISO.</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779076#M1286</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;I am solving &lt;A&gt;&lt;X&gt; = &lt;B&gt; where both X and B are matrices. I 
am solving for X. Size of &lt;/B&gt;&lt;/X&gt;&lt;/A&gt;&lt;A&gt; is 1,500,000 x 1,500,000. Aproximate non 
zeros are 60,000,000. I am able to run my code for the lower size 
problem. But for this size it does not run.&lt;BR /&gt;&lt;BR /&gt;System configuration:&lt;BR /&gt;&lt;BR /&gt;SGI - 8 processor with 32 GB memory.&lt;BR /&gt;Compiler - GNU.&lt;BR /&gt;&lt;BR /&gt;I am using 10.0.1.014 intel mkl&lt;BR /&gt;&lt;BR /&gt;When I am using iparm(60) = 2 it kill the process with -9 error&lt;BR /&gt;&lt;BR /&gt;***************************FOLLOWING WAS ERROR******************************&lt;BR /&gt;&lt;BR /&gt;The file ./pardiso_ooc.cfg was not opened&lt;BR /&gt;*** error PARDISO ( not enough mem. for OOC) error_num= -9&lt;BR /&gt;&lt;BR /&gt;================ PARDISO: solving a symmetric indef. system ================&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;Summary PARDISO: ( reorder to reorder )&lt;BR /&gt;================&lt;BR /&gt;&lt;BR /&gt;Times:&lt;BR /&gt;======&lt;BR /&gt;&lt;BR /&gt; Time fulladj: 1.360909 s&lt;BR /&gt; Time reorder: 29.476320 s&lt;BR /&gt; Time symbfct: 34.058441 s&lt;BR /&gt; Time malloc : 0.000549 s&lt;BR /&gt; Time total : 64.896383 s total - sum: 0.000164 s&lt;BR /&gt;&lt;BR /&gt;Statistics:&lt;BR /&gt;===========&lt;BR /&gt;&amp;lt; Parallel Direct Factorization with #processors: &amp;gt; 1&lt;BR /&gt;&amp;lt; Numerical Factorization with Level-3 BLAS performance &amp;gt;&lt;BR /&gt;&lt;BR /&gt;&amp;lt; Linear system Ax = b&amp;gt; &lt;TRANSPOSE&gt;&lt;BR /&gt; #equations: 1460346&lt;BR /&gt; #non-zeros in A: 58022798&lt;BR /&gt; non-zeros in A (%): 0.002721&lt;BR /&gt; #right-hand sides: 1&lt;BR /&gt;&lt;BR /&gt;&amp;lt; Factors L and U &amp;gt;&lt;BR /&gt; #columns for each panel: 10&lt;BR /&gt; #independent subgraphs: 0&lt;BR /&gt;&amp;lt; Preprocessing with state of the art partitioning metis&amp;gt;&lt;BR /&gt; #supernodes: 231589&lt;BR /&gt; size of largest supernode: 20746&lt;BR /&gt; number of nonzeros in L -905604445&lt;BR /&gt; number of nonzeros in U 1&lt;BR /&gt; number of nonzeros in L+U -905604444&lt;BR /&gt;test&lt;BR /&gt;check point 1&lt;BR /&gt;check point 2&lt;BR /&gt;check point 3&lt;BR /&gt;check point 5&lt;BR /&gt;check point 7&lt;BR /&gt; 58022799&lt;BR /&gt;Reordering completed ... &lt;BR /&gt;The following ERROR was detected: -9&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;BR /&gt;*****************************************************************************&lt;BR /&gt;Questions:&lt;BR /&gt;&lt;BR /&gt;1) Is it possible to solve using current resources (8-Processors with each of 4GB RAM)?&lt;BR /&gt;2)
 If I want to run this size problem then what is the maximum RAM on 
single processor will I need? As we are planning to buy a system that 
has sufficient memory for this type of operations.&lt;BR /&gt;&lt;BR /&gt;Thank you,&lt;BR /&gt;&lt;BR /&gt;Mital Mistry.&lt;/TRANSPOSE&gt;&lt;/A&gt;</description>
      <pubDate>Mon, 01 Aug 2011 14:13:20 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779076#M1286</guid>
      <dc:creator>Mistry__Mital</dc:creator>
      <dc:date>2011-08-01T14:13:20Z</dc:date>
    </item>
    <item>
      <title>Large Scale Problem using PARDISO.</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779077#M1287</link>
      <description>Mital,&lt;BR /&gt;&lt;BR /&gt;I will transfer this issue the MKL User Forum.&lt;BR /&gt;&lt;BR /&gt;ron</description>
      <pubDate>Mon, 01 Aug 2011 15:03:23 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779077#M1287</guid>
      <dc:creator>Ron_Green</dc:creator>
      <dc:date>2011-08-01T15:03:23Z</dc:date>
    </item>
    <item>
      <title>Large Scale Problem using PARDISO.</title>
      <link>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779078#M1288</link>
      <description>Thank you Ron,&lt;BR /&gt;&lt;BR /&gt;I tried by doing,&lt;BR /&gt;&lt;BR /&gt;export MKL_PARDISO_OOC_MAX_CORE_SIZE=8000&lt;BR /&gt;&lt;BR /&gt;And the process did not stop but it kept running for almost 2 days and did nothing. &lt;BR /&gt;&lt;BR /&gt;Also, the process is not possible by in-core right?&lt;BR /&gt;&lt;BR /&gt;Thanks,&lt;BR /&gt;&lt;BR /&gt;Mital&lt;BR /&gt;</description>
      <pubDate>Mon, 01 Aug 2011 15:04:18 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Large-Scale-Problem-using-PARDISO/m-p/779078#M1288</guid>
      <dc:creator>Mistry__Mital</dc:creator>
      <dc:date>2011-08-01T15:04:18Z</dc:date>
    </item>
  </channel>
</rss>

