<?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 &amp;amp; LSF compatibility in Intel® MPI Library</title>
    <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1209734#M7147</link>
    <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;Hello,&lt;/SPAN&gt; &lt;SPAN title=""&gt;Everyone.&lt;/SPAN&gt;&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;There is an issue in which calculation results are different if a specific node is included during forecast model execution based MPI communication.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;I think it's probably the error associated with mpirun's hostfile option.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN class="" title=""&gt;Because, if the hostfile allocated by the LSF job scheduler is executed as it is, no problem occurs.&lt;/SPAN&gt; &lt;SPAN class="" title=""&gt;However, if the hostfile is sorted and used with the -f option, the calculation result is different randomly.&lt;/SPAN&gt;&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;I think this is because the head allocated by the LSF and the head set in the sorted hostfile are different.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;I am wondering if the following two variables can be specified when running mpirun.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;BR /&gt;&lt;SPAN class="" title=""&gt;1) hydra_bstrap_proxy --upstream-host &lt;STRONG&gt;duru0333&lt;/STRONG&gt;&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;2) pmi_proxy --control-port &lt;STRONG&gt;duru0252:44774&lt;/STRONG&gt;&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;I'd like to test if this fits well and the problem does not occur.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thank you in advance.&lt;BR /&gt;Kihang&lt;/P&gt;</description>
    <pubDate>Wed, 16 Sep 2020 01:46:54 GMT</pubDate>
    <dc:creator>youn__kihang</dc:creator>
    <dc:date>2020-09-16T01:46:54Z</dc:date>
    <item>
      <title>Intel MPI &amp; LSF compatibility</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1209734#M7147</link>
      <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;Hello,&lt;/SPAN&gt; &lt;SPAN title=""&gt;Everyone.&lt;/SPAN&gt;&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;There is an issue in which calculation results are different if a specific node is included during forecast model execution based MPI communication.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;I think it's probably the error associated with mpirun's hostfile option.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN class="" title=""&gt;Because, if the hostfile allocated by the LSF job scheduler is executed as it is, no problem occurs.&lt;/SPAN&gt; &lt;SPAN class="" title=""&gt;However, if the hostfile is sorted and used with the -f option, the calculation result is different randomly.&lt;/SPAN&gt;&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;I think this is because the head allocated by the LSF and the head set in the sorted hostfile are different.&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;I am wondering if the following two variables can be specified when running mpirun.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;BR /&gt;&lt;SPAN class="" title=""&gt;1) hydra_bstrap_proxy --upstream-host &lt;STRONG&gt;duru0333&lt;/STRONG&gt;&lt;/SPAN&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;2) pmi_proxy --control-port &lt;STRONG&gt;duru0252:44774&lt;/STRONG&gt;&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;I'd like to test if this fits well and the problem does not occur.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;Thank you in advance.&lt;BR /&gt;Kihang&lt;/P&gt;</description>
      <pubDate>Wed, 16 Sep 2020 01:46:54 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1209734#M7147</guid>
      <dc:creator>youn__kihang</dc:creator>
      <dc:date>2020-09-16T01:46:54Z</dc:date>
    </item>
    <item>
      <title>Re:Intel MPI &amp;amp; LSF compatibility</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1209844#M7153</link>
      <description>&lt;P&gt;Hi Kihang,&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Does this issue occur if the specific node is not included in the queue?&lt;/P&gt;&lt;P&gt;How are you sorting the node list? and why?&lt;/P&gt;&lt;P&gt;Can you set &lt;I&gt;I_MPI_HYDRA_DEBUG=on&lt;/I&gt; and see if &lt;B&gt;-&lt;/B&gt;&lt;I&gt;-upstream-host&lt;/I&gt;&amp;nbsp;is different in both the cases? &lt;/P&gt;&lt;P&gt;(1. Using default nodelist in $PBS_NODEFILE &lt;/P&gt;&lt;P&gt;2. The sorted hostfile you have generated)&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;In our case, node is same as in PBS_NODEFILE despite the change of&amp;nbsp;order in hostfile.&lt;/P&gt;&lt;P&gt;Can you provide us with any reproducer code?&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Regarding setting the upstream host manually we will discuss with our internal team and get back to you.&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Regards&lt;/P&gt;&lt;P&gt;Prasanth&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Wed, 16 Sep 2020 12:36:55 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1209844#M7153</guid>
      <dc:creator>PrasanthD_intel</dc:creator>
      <dc:date>2020-09-16T12:36:55Z</dc:date>
    </item>
    <item>
      <title>Re: Re:Intel MPI &amp;amp; LSF compatibility</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1210036#M7157</link>
      <description>&lt;P&gt;&amp;nbsp;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;Hello Prasanth,&lt;/SPAN&gt;&lt;BR /&gt;&lt;BR /&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;The error I mentioned in the previous post was a mistake in identifying the cause.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;The error did not occur according to the order of the host.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN class="" title=""&gt;It happened even when running on one node, and the cause is also different, so I will close it here and open another post.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;BR /&gt;&lt;SPAN title=""&gt;Thank you.&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;
&lt;P&gt;&lt;SPAN class="tlid-translation translation"&gt;&lt;SPAN title=""&gt;Kihang&lt;/SPAN&gt;&lt;/SPAN&gt;&lt;/P&gt;</description>
      <pubDate>Thu, 17 Sep 2020 01:52:22 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1210036#M7157</guid>
      <dc:creator>youn__kihang</dc:creator>
      <dc:date>2020-09-17T01:52:22Z</dc:date>
    </item>
    <item>
      <title>Re:Intel MPI &amp;amp; LSF compatibility</title>
      <link>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1211122#M7172</link>
      <description>&lt;P&gt;Hi Kihang,&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Since you have raised a new thread for your problem and as you have suggested, we are closing this thread.&lt;/P&gt;&lt;P&gt;Any further interaction in this thread will be considered community only&lt;/P&gt;&lt;P&gt;&lt;BR /&gt;&lt;/P&gt;&lt;P&gt;Regards&lt;/P&gt;&lt;P&gt;Prasanth&lt;/P&gt;&lt;BR /&gt;</description>
      <pubDate>Mon, 21 Sep 2020 10:52:33 GMT</pubDate>
      <guid>https://community.intel.com/t5/Intel-MPI-Library/Intel-MPI-amp-LSF-compatibility/m-p/1211122#M7172</guid>
      <dc:creator>PrasanthD_intel</dc:creator>
      <dc:date>2020-09-21T10:52:33Z</dc:date>
    </item>
  </channel>
</rss>

