<?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 Intel MPI and Infiniband: MTU optimisation doc ? in Intel® MPI Library</title>
    <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771527#M190</link>
    <description>Hi Guillaume,&lt;BR /&gt;&lt;BR /&gt;It seems to me that you'd better ask this question on a forum dedicated to CFD, might be people from MTU can help you. Here you rather ask engineers from Intel familiar with an Intel product.&lt;BR /&gt;&lt;BR /&gt;You can use Intel MPI Library with default settings - that should be OK. But to be sure that you are using fast fabrics you can set I_MPI_FALLBACK=0 and I_MPI_FABRICS=shm:dapl&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt; Dmitry&lt;BR /&gt;</description>
    <pubDate>Sat, 04 Dec 2010 09:25:34 GMT</pubDate>
    <dc:creator>Dmitry_K_Intel2</dc:creator>
    <dc:date>2010-12-04T09:25:34Z</dc:date>
    <item>
      <title>Intel MPI and Infiniband: MTU optimisation doc ?</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771526#M189</link>
      <description>Hi,&lt;BR /&gt;&lt;BR /&gt;We have a little cluster:&lt;BR /&gt;- master&lt;BR /&gt;- 8 nodes with 12 cores pro nodes&lt;BR /&gt;- Infiniband connectx DDR 2X&lt;BR /&gt;- linux CentOS 5.5&lt;BR /&gt;- Infiniband stack from OFED&lt;BR /&gt;- Intel MPI&lt;BR /&gt;&lt;BR /&gt;At the moment we have a normal basic install of the infiniband packages. So we are in connected mode with a MTU of ~65k. Has anybody info, documentation, links on the optimization of the MTU ? We are using it for Computational Fluid Dynamics computations.&lt;BR /&gt;&lt;BR /&gt;Thx in advance,&lt;BR /&gt;Best regards,&lt;BR /&gt;Guillaume</description>
      <pubDate>Sat, 04 Dec 2010 08:40:26 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771526#M189</guid>
      <dc:creator>Guillaume_De_Nayer</dc:creator>
      <dc:date>2010-12-04T08:40:26Z</dc:date>
    </item>
    <item>
      <title>Intel MPI and Infiniband: MTU optimisation doc ?</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771527#M190</link>
      <description>Hi Guillaume,&lt;BR /&gt;&lt;BR /&gt;It seems to me that you'd better ask this question on a forum dedicated to CFD, might be people from MTU can help you. Here you rather ask engineers from Intel familiar with an Intel product.&lt;BR /&gt;&lt;BR /&gt;You can use Intel MPI Library with default settings - that should be OK. But to be sure that you are using fast fabrics you can set I_MPI_FALLBACK=0 and I_MPI_FABRICS=shm:dapl&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt; Dmitry&lt;BR /&gt;</description>
      <pubDate>Sat, 04 Dec 2010 09:25:34 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771527#M190</guid>
      <dc:creator>Dmitry_K_Intel2</dc:creator>
      <dc:date>2010-12-04T09:25:34Z</dc:date>
    </item>
    <item>
      <title>Intel MPI and Infiniband: MTU optimisation doc ?</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771528#M191</link>
      <description>Hi Dmitry,&lt;BR /&gt;&lt;BR /&gt;A lot of CFD engineers are using intel mpi...so perhaps one or two will look at this thread ;)&lt;BR /&gt;&lt;BR /&gt;thx for your answer.&lt;BR /&gt;regards&lt;BR /&gt;Guillaume&lt;BR /&gt;</description>
      <pubDate>Sat, 04 Dec 2010 10:24:07 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771528#M191</guid>
      <dc:creator>Guillaume_De_Nayer</dc:creator>
      <dc:date>2010-12-04T10:24:07Z</dc:date>
    </item>
    <item>
      <title>Intel MPI and Infiniband: MTU optimisation doc ?</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771529#M192</link>
      <description>Hi Dmitry,&lt;BR /&gt;&lt;BR /&gt;I have another question: I'm suing the OFED stack for infiniband: Which I_MPI_FABRICS should I choose ? shm:dapl or shm:ofa ?&lt;BR /&gt;&lt;BR /&gt;Thx,&lt;BR /&gt;best regards,&lt;BR /&gt;Guillaume&lt;BR /&gt;</description>
      <pubDate>Mon, 06 Dec 2010 08:51:21 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771529#M192</guid>
      <dc:creator>Guillaume_De_Nayer</dc:creator>
      <dc:date>2010-12-06T08:51:21Z</dc:date>
    </item>
    <item>
      <title>Intel MPI and Infiniband: MTU optimisation doc ?</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771530#M193</link>
      <description>Hi Guillaume,&lt;BR /&gt;&lt;BR /&gt;You can use any.&lt;BR /&gt;OFA fabric gives you ability to use multi-rail feature (if you have 2 IB cards on a node or cards with 2 ports). Just compare performance of dapl and ofa with your application and use the fastest.&lt;BR /&gt;&lt;BR /&gt;Regards!&lt;BR /&gt; Dmitry&lt;BR /&gt;</description>
      <pubDate>Mon, 06 Dec 2010 08:57:59 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-and-Infiniband-MTU-optimisation-doc/m-p/771530#M193</guid>
      <dc:creator>Dmitry_K_Intel2</dc:creator>
      <dc:date>2010-12-06T08:57:59Z</dc:date>
    </item>
  </channel>
</rss>

