<div dir="ltr">Phil (of Lustre fame) captures some of it here...<div><br></div><div><a href="https://www.hpcwire.com/2016/03/23/downunder-geosolution-leverages-xeon-phi-improved-seismic-processing/">https://www.hpcwire.com/2016/03/23/downunder-geosolution-leverages-xeon-phi-improved-seismic-processing/</a><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Aug 19, 2016 at 10:30 PM, Prentice Bisbal <span dir="ltr"><<a href="mailto:pbisbal@pppl.gov" target="_blank">pbisbal@pppl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
  
    
  
  <div bgcolor="#FFFFFF" text="#000000">
    <p>I'm not going to dispute your claims, but I think you're probably
      a statistical anomaly in this regard. I'm guessing your code isn't
      that old, so it was built using current 'best practices' for
      parallelization (loop independence, data parallelism, overlapping
      communication and computation, etc) for the get-go, or was at
      least kept 'modern'  which made it easy for the compilers to
      optimize for the Phis. <br>
    </p>
    Can you comment on this? <br><span class="HOEnZb"><font color="#888888">
    <pre cols="72">Prentice</pre></font></span><div><div class="h5">
    <div>On 08/18/2016 09:28 PM, Stu Midgley
      wrote:<br>
    </div>
    <blockquote type="cite">
      <div dir="ltr">we get EXCELLENT performance from the phi's... just
        look at our marketing with Intel.
        <div><br>
        </div>
        <div>The 72XX series are also superb.</div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Fri, Aug 19, 2016 at 7:41 AM,
          Christopher Samuel <span dir="ltr"><<a href="mailto:samuel@unimelb.edu.au" target="_blank">samuel@unimelb.edu.au</a>></span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><span>On 18/08/16 00:10, Prentice Bisbal wrote:<br>
              <br>
              > I wouldn't be so glad. You're still going to have to
              rewrite your code<br>
              > as mentioned above to get any meaningful performance.<br>
              <br>
            </span>To be fair to Stu though he did post back in 2013
            that he got good<br>
            performance improvements on his codes on Xeon Phi for little
            initial<br>
            effort, so that's why they're so attractive to him.<br>
            <br>
            <a href="http://www.beowulf.org/pipermail/beowulf/2013-February/030963.html" rel="noreferrer" target="_blank">http://www.beowulf.org/piperma<wbr>il/beowulf/2013-February/<wbr>030963.html</a><br>
            <br>
            <a href="http://www.beowulf.org/pipermail/beowulf/2013-February/030971.html" rel="noreferrer" target="_blank">http://www.beowulf.org/piperma<wbr>il/beowulf/2013-February/<wbr>030971.html</a><br>
            <br>
            To us though they've proved not that useful, we can only run
            them<br>
            in accelerator mode (due to external constraints) and so
            we've had<br>
            virtually 0 uptake. :-(<br>
            <br>
            All the best,<br>
            Chris<br>
            <span><font color="#888888">--<br>
                 Christopher Samuel        Senior Systems Administrator<br>
                 VLSCI - Victorian Life Sciences Computation Initiative<br>
                 Email: <a href="mailto:samuel@unimelb.edu.au" target="_blank">samuel@unimelb.edu.au</a>
                Phone: <a href="tel:%2B61%20%280%293%20903%2055545" value="+61390355545" target="_blank">+61 (0)3 903 55545</a><br>
                 <a href="http://www.vlsci.org.au/" rel="noreferrer" target="_blank">http://www.vlsci.org.au/</a>      <a href="http://twitter.com/vlsci" rel="noreferrer" target="_blank">http://twitter.com/vlsci</a><br>
              </font></span>
            <div>
              <div>______________________________<wbr>_________________<br>
                Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a>
                sponsored by Penguin Computing<br>
                To change your subscription (digest mode or unsubscribe)
                visit <a href="http://www.beowulf.org/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">http://www.beowulf.org/mailman<wbr>/listinfo/beowulf</a><br>
              </div>
            </div>
          </blockquote>
        </div>
        <br>
        <br clear="all">
        <div><br>
        </div>
        -- <br>
        <div data-smartmail="gmail_signature">Dr
          Stuart Midgley<br>
          <a href="mailto:sdm900@sdm900.com" target="_blank">sdm900@sdm900.com</a></div>
      </div>
      <br>
      <fieldset></fieldset>
      <br>
      <pre>______________________________<wbr>_________________
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org" target="_blank">Beowulf@beowulf.org</a> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit <a href="http://www.beowulf.org/mailman/listinfo/beowulf" target="_blank">http://www.beowulf.org/<wbr>mailman/listinfo/beowulf</a>
</pre>
    </blockquote>
    <br>
  </div></div></div>

<br>______________________________<wbr>_________________<br>
Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a> sponsored by Penguin Computing<br>
To change your subscription (digest mode or unsubscribe) visit <a href="http://www.beowulf.org/mailman/listinfo/beowulf" rel="noreferrer" target="_blank">http://www.beowulf.org/<wbr>mailman/listinfo/beowulf</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature" data-smartmail="gmail_signature">Dr Stuart Midgley<br><a href="mailto:sdm900@sdm900.com" target="_blank">sdm900@sdm900.com</a></div>
</div>