<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    Hi Yaoping,<br>
    <br>
    What parallelization option did you compile WRF with (e.g. MPI
    only)? Also, I've found the Intel compilers to be 3 times faster
    than the GNU compilers on some WRF configurations (unfortunately).
    What compiler did you use?<br>
    <br>
    You may also want to experiment using less than the number of total
    cores on the machine. For example, you could try using 44 cores
    instead of 48. I think WRF EMS is set to do this by default. I've
    verified on some of my multi-core machines that this does indeed
    reduce the runtime.<br>
    <br>
    Cheers,<br>
    Mike<br>
    <br>
    <br>
    <div class="moz-cite-prefix">On 09/01/2016 03:15 PM, Carlos Ross
      wrote:<br>
    </div>
    <blockquote
cite="mid:CAMCa-Quf8n9EDb7ZpxtRU_RD9ABquyrdpTquW0sD8LfB=n-_ng@mail.gmail.com"
      type="cite">
      <div dir="ltr">I think it should be faster, <span
style="color:rgb(0,0,0);font-family:mangal,sans-serif;font-size:13.3333px">Xeon
          x5650 CPUs are 6 cores and 12 threads, so you maybe using 2
          process in one core and that is slowing it down.</span></div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">2016-08-31 18:36 GMT-03:00 Wang,
          Yaoping <span dir="ltr">&lt;<a moz-do-not-send="true"
              href="mailto:wang.3866@buckeyemail.osu.edu"
              target="_blank">wang.3866@buckeyemail.osu.edu</a>&gt;</span>:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div dir="ltr">
              <div
style="font-size:10pt;color:#000000;background-color:#ffffff;font-family:Mangal,Sans-Serif">
                <p>Hi All, </p>
                <p><br>
                </p>
                <p>I am running WRF on a ~6km resolution, <span
                    style="font-size:10pt">91 x 121 domain in the
                    eastern United States. I am using an adaptive time
                    step which makes it mostly 72 sec increments. </span><span
                    style="font-size:10pt">There are 34 vertical
                    levels. I use 4 x 12 cores on a Intel Xeon x5650
                    CPUs machine. </span><span style="font-size:10pt">The
                    throughput is about 1.2 hour wall time per 24 hours
                    model time. </span></p>
                <p><span style="font-size:10pt"></span></p>
                <p><span style="font-size:10pt"><br>
                  </span></p>
                <p><span style="font-size:10pt">Is this a reasonable
                    speed? I found some information here (<a
                      moz-do-not-send="true"
href="http://www.ecmwf.int/sites/default/files/elibrary/2014/13662-performance-analysis-operational-implementaion-wrf.pdf"
                      target="_blank">http://www.ecmwf.int/sites/<wbr>default/files/elibrary/2014/<wbr>13662-performance-analysis-<wbr>operational-implementaion-wrf.<wbr>pdf</a>)
                    and after considering the domain difference, my run
                    still seems a touch slow. And i</span><span
                    style="font-size:10pt">s there anyway I could figure
                    how to make the model run faster?</span></p>
                <p><span style="font-size:10pt"><br>
                  </span></p>
                <p><span style="font-size:10pt">Thank you, </span></p>
                <span class="HOEnZb"><font color="#888888">
                    <p><span style="font-size:10pt">Yaoping Wang</span></p>
                  </font></span></div>
            </div>
            <br>
            ______________________________<wbr>_________________<br>
            Wrf-users mailing list<br>
            <a moz-do-not-send="true" href="mailto:Wrf-users@ucar.edu">Wrf-users@ucar.edu</a><br>
            <a moz-do-not-send="true"
              href="http://mailman.ucar.edu/mailman/listinfo/wrf-users"
              rel="noreferrer" target="_blank">http://mailman.ucar.edu/<wbr>mailman/listinfo/wrf-users</a><br>
            <br>
          </blockquote>
        </div>
        <br>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Wrf-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Wrf-users@ucar.edu">Wrf-users@ucar.edu</a>
<a class="moz-txt-link-freetext" href="http://mailman.ucar.edu/mailman/listinfo/wrf-users">http://mailman.ucar.edu/mailman/listinfo/wrf-users</a>
</pre>
    </blockquote>
    <br>
    <div class="moz-signature">-- <br>
      <table border="0" cellpadding="2" cellspacing="2" height="126"
        width="15%">
        <tbody>
          <tr>
            <td valign="center"><img alt="Sail Tactics logo"
                src="cid:part5.FD4A0143.10B55550@sailtactics.com"
                height="118" width="90"><br>
            </td>
            <td valign="top">Mike Dvorak, PhD<br>
              Founder<br>
              Sail Tactics, LLC<br>
              Corpus Christi, TX<br>
              +1 650-454-5243<br>
              <a href="http://sailtactics.com">http://sailtactics.com</a><br>
            </td>
          </tr>
        </tbody>
      </table>
    </div>
  </body>
</html>