<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>I remember in 1999/2000 I did a simple BLAST benchmark on my SGI
      Pentium workstation running linux, and one on my R10k box. 
      Similar clock speeds.  Pentium was 2x faster in Int heavy calcs. 
      Did similar things with FP, and discussed the results with my team
      and engineering.  <br>
    </p>
    <p>None of them were happy.  <br>
    </p>
    <p>I left SGI in 2001, after a discussion with the SVP of
      engineering, Warren Pratt, who insisted that the future was not
      linux clusters, but large beefy many cpu shared memory servers.  I
      put my money/future where my mouth was.</p>
    <p>Today, I run on rackmount 1 and 2U boxes containing 80-128 CPU
      (cores), and 2TB physical ram.</p>
    <p>So yeah.  He was right.</p>
    <p>But they run linux, and are x86_64 arch.<br>
    </p>
    <p><br>
    </p>
    <div class="moz-cite-prefix">On 3/23/23 15:08, Prentice Bisbal via
      Beowulf wrote:<br>
    </div>
    <blockquote type="cite"
      cite="mid:b1c305a5-e445-6713-19d4-d54bf3e66a42@pppl.gov">
      <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
      <p>Yeah, that whole situation was frustrating. From 2004 -2007, I
        was working for a pharmaceutical startup supporting their
        Computer-Aided Drug Discovery (CADD) team. Had I been hired
        before the director of CADD, it would have been a 100% Linux
        shop. Instead, as soon as he was hired he started insisting, and
        circulated a memo, stating that Linux was still a toy for
        hobbyists and "not ready for primetime" (he used that exact
        quote). So we spent tens of thousands of dollars on two Octanes
        and the 8-way Origin 350. I got a Linux workstation as a
        proof-of-concept, and that HP Workstation running Linux that
        cost only a few thousand dollars ran circles around those SGI
        boxes, and when cost was factored in FLOPS/$ was like 10x better
        than the SGI hardware at that point. And all of that hardware
        was bought used ("remarketed") from SGI, so new hardware would
        have compared significantly worse in terms of value. <br>
      </p>
      <p>Also, it turned out the director of CADD owned a nontrivial
        amount of SGI stocks, so not only was an he over-the-hill
        curmudgeon afraid of new technology, there was also a pretty
        clear conflict of interest for him to be pushing SGI, even
        though I'm sure our small purchase did nothing to improve SGI
        stock value. <br>
      </p>
      <div class="moz-cite-prefix">On 3/23/23 2:58 PM, Joe Landman
        wrote:<br>
      </div>
      <blockquote type="cite"
        cite="mid:bc99ae44-5ce7-28b7-c9fb-b686cf4d22df@gmail.com">
        <meta http-equiv="Content-Type" content="text/html;
          charset=UTF-8">
        <p>They had laid off all the good people doing workstations by
          then, I think they outsourced design/production to ODMs by
          that time.  MIP processors were long in the tooth in 1999,
          never mind 2007.</p>
        <p>Having been at SGI from 1995-2001, I can tell you the reason
          MIPS sucked wind at that point, was the good ship Itanic sunk
          Alien and Beast processors.  Those design teams left, and we
          didn't have much for post R10k, other than respins and shrinks
          of R10k.  Which were renamed R12k, R14k ...</p>
        <p>Beast would have been relevant (near EOL though) in 2007.  <br>
        </p>
        <p><br>
        </p>
        <div class="moz-cite-prefix">On 3/23/23 14:53, Prentice Bisbal
          via Beowulf wrote:<br>
        </div>
        <blockquote type="cite"
          cite="mid:4f14ef2e-afdf-0e52-e1c5-0ad47b7afcc8@pppl.gov">
          <meta http-equiv="Content-Type" content="text/html;
            charset=UTF-8">
          <p>Between 2003 and 2007, I worked with a lot of O2s, Octanes,
            an 8-way Origin 350, and even a Tezro. I don't miss those
            days. <br>
          </p>
          <p>I always felt like the design of their workstations was
            done by the same people who design Playskool toys rather
            than professional hardware. <br>
          </p>
          <pre class="moz-signature" cols="72">Prentice Bisbal
Senior HPC Engineer
Computational Sciences Department
Princeton Plasma Physics Laboratory
Princeton, NJ
<a class="moz-txt-link-freetext" href="https://cs.pppl.gov" moz-do-not-send="true">https://cs.pppl.gov</a>
<a class="moz-txt-link-freetext" href="https://www.pppl.gov" moz-do-not-send="true">https://www.pppl.gov</a></pre>
          <div class="moz-cite-prefix">On 3/23/23 1:08 PM, Ryan
            Novosielski via Beowulf wrote:<br>
          </div>
          <blockquote type="cite"
            cite="mid:87C8A045-0386-45E1-AD8A-4ACB952BE1FD@rutgers.edu">
            <meta http-equiv="Content-Type" content="text/html;
              charset=UTF-8">
            Seriously. I have an Indy and an Octane2 laying around.
            That’s not even an SGI. :-P
            <div><br>
              <div>
                <div dir="auto" style="caret-color: rgb(0, 0, 0); color:
                  rgb(0, 0, 0); letter-spacing: normal; text-align:
                  start; text-indent: 0px; text-transform: none;
                  white-space: normal; word-spacing: 0px;
                  -webkit-text-stroke-width: 0px; text-decoration: none;
                  overflow-wrap: break-word; -webkit-nbsp-mode: space;
                  line-break: after-white-space;">
                  <div style="color: rgb(0, 0, 0); letter-spacing:
                    normal; text-align: start; text-indent: 0px;
                    text-transform: none; white-space: normal;
                    word-spacing: 0px; -webkit-text-stroke-width: 0px;
                    overflow-wrap: break-word; -webkit-nbsp-mode: space;
                    line-break: after-white-space;"> --<br>
                    #BlackLivesMatter</div>
                  <div style="color: rgb(0, 0, 0); letter-spacing:
                    normal; text-align: start; text-indent: 0px;
                    text-transform: none; white-space: normal;
                    word-spacing: 0px; -webkit-text-stroke-width: 0px;
                    overflow-wrap: break-word; -webkit-nbsp-mode: space;
                    line-break: after-white-space;"> ____<br>
                    || \\UTGERS,  <span class="Apple-tab-span" style="white-space: pre;"> </span> |---------------------------*O*---------------------------<br>
                    ||_// the State<span class="Apple-tab-span" style="white-space: pre;"> </span> |
                            Ryan Novosielski - <a
                      class="moz-txt-link-abbreviated
                      moz-txt-link-freetext"
                      href="mailto:novosirj@rutgers.edu"
                      moz-do-not-send="true">novosirj@rutgers.edu</a><br>
                    || \\ University | Sr. Technologist - 973/972.0922
                    (2x0922) ~*~ RBHS Campus<br>
                    ||  \\    of NJ<span class="Apple-tab-span" style="white-space: pre;"> </span> |
                    Office of Advanced Research Computing - MSB
                    A555B, Newark<br>
                         `'</div>
                </div>
              </div>
              <div><br>
                <blockquote type="cite">
                  <div>On Mar 23, 2023, at 13:07, Michael DiDomenico <a
                      class="moz-txt-link-rfc2396E"
                      href="mailto:mdidomenico4@gmail.com"
                      moz-do-not-send="true"><mdidomenico4@gmail.com></a>
                    wrote:</div>
                  <br class="Apple-interchange-newline">
                  <div>
                    <div>ack, irix flashbacks... :) fortunately, this
                      machine isn't quite that<br>
                      old, circa 2013<br>
                      <br>
                      On Thu, Mar 23, 2023 at 1:02 PM Darren Wise via
                      Beowulf<br>
                      <a class="moz-txt-link-rfc2396E"
                        href="mailto:beowulf@beowulf.org"
                        moz-do-not-send="true"><beowulf@beowulf.org></a>
                      wrote:<br>
                      <blockquote type="cite"><br>
                        Hello,<br>
                        <br>
                        I don't personally have such myself but anything
                        SGI even with today's<br>
                        exotics Ians about the most knowledgable I know
                        of globally and<br>
                        seriously dying out within this sector, I hazard
                        a guess your SM board<br>
                        would be considered quite new compared to other
                        systems.<br>
                        <br>
                        <a class="moz-txt-link-freetext"
                          href="http://www.sgidepot.co.uk/sgidepot/"
                          moz-do-not-send="true">http://www.sgidepot.co.uk/sgidepot/</a><br>
                        <br>
                        <br>
                        Kind Regards,<br>
                        Darren Wise<br>
                        Research Engineer, Mechatronics<br>
                        <a class="moz-txt-link-freetext"
                          href="https://wisecorp.co.uk"
                          moz-do-not-send="true">https://wisecorp.co.uk</a>,
                        .us & .ru<br>
                        <br>
                        On 23/03/2023 16:51, Michael DiDomenico wrote:<br>
                        <blockquote type="cite">does anyone happen to
                          have an old sgi / supermicro bios for an<br>
                          X9DRG-QF+ motherboard squirreled away
                          somewhere?  sgi is long gone,<br>
                          hpe might have something still but who knows
                          where.  i reached out to<br>
                          supermicro, but i suspect they'll say no.<br>
_______________________________________________<br>
                          Beowulf mailing list, <a
                            class="moz-txt-link-abbreviated
                            moz-txt-link-freetext"
                            href="mailto:Beowulf@beowulf.org"
                            moz-do-not-send="true">Beowulf@beowulf.org</a>
                          sponsored by Penguin Computing<br>
                          To change your subscription (digest mode or
                          unsubscribe) visit <a
                            class="moz-txt-link-freetext"
                            href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf"
                            moz-do-not-send="true">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
                        </blockquote>
                        _______________________________________________<br>
                        Beowulf mailing list, <a
                          class="moz-txt-link-abbreviated
                          moz-txt-link-freetext"
                          href="mailto:Beowulf@beowulf.org"
                          moz-do-not-send="true">Beowulf@beowulf.org</a>
                        sponsored by Penguin Computing<br>
                        To change your subscription (digest mode or
                        unsubscribe) visit <a
                          class="moz-txt-link-freetext"
                          href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf"
                          moz-do-not-send="true">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
                      </blockquote>
                      _______________________________________________<br>
                      Beowulf mailing list, <a
                        class="moz-txt-link-abbreviated
                        moz-txt-link-freetext"
                        href="mailto:Beowulf@beowulf.org"
                        moz-do-not-send="true">Beowulf@beowulf.org</a>
                      sponsored by Penguin Computing<br>
                      To change your subscription (digest mode or
                      unsubscribe) visit <a
                        class="moz-txt-link-freetext"
                        href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf"
                        moz-do-not-send="true">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a><br>
                    </div>
                  </div>
                </blockquote>
              </div>
              <br>
            </div>
            <br>
            <fieldset class="moz-mime-attachment-header"></fieldset>
            <pre class="moz-quote-pre" wrap="">_______________________________________________
Beowulf mailing list, <a class="moz-txt-link-abbreviated moz-txt-link-freetext" href="mailto:Beowulf@beowulf.org" moz-do-not-send="true">Beowulf@beowulf.org</a> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit <a class="moz-txt-link-freetext" href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" moz-do-not-send="true">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a>
</pre>
          </blockquote>
          <br>
          <fieldset class="moz-mime-attachment-header"></fieldset>
          <pre class="moz-quote-pre" wrap="">_______________________________________________
Beowulf mailing list, <a class="moz-txt-link-abbreviated moz-txt-link-freetext" href="mailto:Beowulf@beowulf.org" moz-do-not-send="true">Beowulf@beowulf.org</a> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit <a class="moz-txt-link-freetext" href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" moz-do-not-send="true">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a>
</pre>
        </blockquote>
        <pre class="moz-signature" cols="72">-- 
Joe Landman
e: <a class="moz-txt-link-abbreviated moz-txt-link-freetext" href="mailto:joe.landman@gmail.com" moz-do-not-send="true">joe.landman@gmail.com</a>
t: @hpcjoe
w: <a class="moz-txt-link-freetext" href="https://scalability.org" moz-do-not-send="true">https://scalability.org</a>
g: <a class="moz-txt-link-freetext" href="https://github.com/joelandman" moz-do-not-send="true">https://github.com/joelandman</a>
l: <a class="moz-txt-link-freetext" href="https://www.linkedin.com/in/joelandman" moz-do-not-send="true">https://www.linkedin.com/in/joelandman</a></pre>
        <br>
        <fieldset class="moz-mime-attachment-header"></fieldset>
        <pre class="moz-quote-pre" wrap="">_______________________________________________
Beowulf mailing list, <a class="moz-txt-link-abbreviated moz-txt-link-freetext" href="mailto:Beowulf@beowulf.org" moz-do-not-send="true">Beowulf@beowulf.org</a> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit <a class="moz-txt-link-freetext" href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf" moz-do-not-send="true">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a>
</pre>
      </blockquote>
      <br>
      <fieldset class="moz-mime-attachment-header"></fieldset>
      <pre class="moz-quote-pre" wrap="">_______________________________________________
Beowulf mailing list, <a class="moz-txt-link-abbreviated" href="mailto:Beowulf@beowulf.org">Beowulf@beowulf.org</a> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit <a class="moz-txt-link-freetext" href="https://beowulf.org/cgi-bin/mailman/listinfo/beowulf">https://beowulf.org/cgi-bin/mailman/listinfo/beowulf</a>
</pre>
    </blockquote>
    <pre class="moz-signature" cols="72">-- 
Joe Landman
e: <a class="moz-txt-link-abbreviated" href="mailto:joe.landman@gmail.com">joe.landman@gmail.com</a>
t: @hpcjoe
w: <a class="moz-txt-link-freetext" href="https://scalability.org">https://scalability.org</a>
g: <a class="moz-txt-link-freetext" href="https://github.com/joelandman">https://github.com/joelandman</a>
l: <a class="moz-txt-link-freetext" href="https://www.linkedin.com/in/joelandman">https://www.linkedin.com/in/joelandman</a></pre>
  </body>
</html>