<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <p>
      <blockquote type="cite"> I suspect though that you need servers
        engineered to fit onto their heatsinks.<br>
      </blockquote>
    </p>
    <p>This is the main crux of my previous criticism for direct-contact
      liquid-cooling solutions, the cooling piping and heat exchanger(s)
      can be retrofit to existing server designs with minimal
      re-engineering. And even for immersive cooling, you just remove
      the case altogether, or make it completely sealed to contain the
      coolant, both of which require minimal re-engineering. This
      solution looks like it would require major rethinking of how
      components are organized. <br>
    </p>
    <pre class="moz-signature" cols="72">Prentice </pre>
    <div class="moz-cite-prefix">On 1/28/19 12:24 PM, John Hearns wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAPqNE2Vc9_OM9XZyDHmJetY2yqVVfnL=XKmV1vhJsUSkP+_VQw@mail.gmail.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="ltr">
        <div dir="ltr">
          <div dir="ltr">
            <div>Prentice, the website refers to Open Compute racks. 
              "... technology has been designed to fit into standard
              Open Compute racks".</div>
            <div>So yep, 19 inch racks are not being targeted here. But
              OCP is pretty widespread.</div>
            <div>I would really like to find out if they can retrofit
              these to existing kit. I suspect though that you need
              servers engineered to fit onto their heatsinks.</div>
            <div><span style="text-align:left;color:rgb(255,255,255);text-transform:none;text-indent:0px;letter-spacing:0px;font-family:myriad-pro;font-size:22px;font-style:normal;font-variant:normal;font-weight:400;text-decoration:none;word-spacing:0px;display:inline;white-space:pre-wrap;word-wrap:break-word;float:none;background-color:transparent"><span style="text-align:left;color:rgb(255,255,255);text-transform:none;text-indent:0px;letter-spacing:0px;font-family:myriad-pro;font-size:22px;font-style:normal;font-variant:normal;font-weight:400;text-decoration:none;word-spacing:0px;display:inline;white-space:pre-wrap;word-wrap:break-word;float:none;background-color:transparent">orced Physics cooling technology has been designed to fit into Standard Open Compute Racks. </span>orced Physics cooling technology has been designed to fit into Standard Open Compute Racks. </span></div>
            <div><br>
            </div>
          </div>
        </div>
      </div>
      <br>
      <div class="gmail_quote">
        <div class="gmail_attr" dir="ltr">On Mon, 28 Jan 2019 at 16:48,
          Prentice Bisbal via Beowulf <<a
            href="mailto:beowulf@beowulf.org" moz-do-not-send="true">beowulf@beowulf.org</a>>
          wrote:<br>
        </div>
        <blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid">
          <div bgcolor="#FFFFFF">
            <p>So I was thinking about this over the weekend (because I
              apparently have nothing better to do with my time), and I
              definitely think this is a non-starter due to the massive
              change in server hardware layout to accommodate this
              thing. Yes, blades, and twin form factor servers already
              required that, and they're common form factors, but those
              form factors were just a matter of shrinking or changing
              the layout of the motherboard but still look like
              "traditional" layouts to the untrained eye, and they were
              still designed with typical front-to-back air cooling in
              mind. I feel like re-arranging the layout of components to
              accomodate this thing is a just more of a change than the
              market will accept. <br>
            </p>
            <p>Just my 2 cents. <br>
            </p>
            <pre class="gmail-m_-4339801130891775252moz-signature" cols="72">Prentice </pre>
            <div class="gmail-m_-4339801130891775252moz-cite-prefix">On
              1/25/19 3:56 PM, Prentice Bisbal wrote:<br>
            </div>
            <blockquote type="cite">
              <p>Eric, <br>
              </p>
              <p>I was suspecting that might be the case, but the
                explanations in the other articles were way too vague to
                be sure of that. The NextPlatform provided much better
                pictures. If that's the case, this thing operates like a
                direct-expansion (DX) refrigeration system, where the
                refrigerant is air and does not change state from liquid
                to gas, like a typical DX refrigeration system, and the
                induced-draft fan provides the shaft work, and those
                tiny channels that allegedly line up the molecules act
                as many tiny offices for the throttling process. Based
                on the pictures in the Next Platform article, here is a
                crude drawing of cross-section of one of these devices
                that I drew in Google Draw. It should help you
                understand what's going inside this thing: <br>
              </p>
              <p><a
                  class="gmail-m_-4339801130891775252moz-txt-link-freetext"
href="https://docs.google.com/document/d/1UK94PxVlQtVSb2ns5TbCqHjPJ1vYSOmkGSeSorvHyaM/edit?usp=sharing"
                  target="_blank" moz-do-not-send="true">https://docs.google.com/document/d/1UK94PxVlQtVSb2ns5TbCqHjPJ1vYSOmkGSeSorvHyaM/edit?usp=sharing</a></p>
              <p>Given this design, you can only have an induced-draft
                fan on the outlet. A forced-draft fan on the inlet would
                compress the air, heating it up and negating the
                throttling (or Joule-Thompson) effect on the
                low-pressure side. <br>
              </p>
              <p>At the end of the day, thermodynamics still says X
                amount of shaft work has to be done to provide Y amount
                of cooling through this process, so I'm still skeptical
                of it, especially at scale. <br>
              </p>
              <p>And for those of you looking for something really
                boring to read rather than work, here are the related
                patents. I haven't read them myself. <br>
              </p>
              <p><a
                  class="gmail-m_-4339801130891775252moz-txt-link-freetext"
                  href="https://patents.google.com/patent/US8414847"
                  target="_blank" moz-do-not-send="true">https://patents.google.com/patent/US8414847</a></p>
              <p><a
                  class="gmail-m_-4339801130891775252moz-txt-link-freetext"
                  href="https://patents.google.com/patent/US8986627B2"
                  target="_blank" moz-do-not-send="true">https://patents.google.com/patent/US8986627B2</a></p>
              <p><a
                  class="gmail-m_-4339801130891775252moz-txt-link-freetext"
                  href="https://patents.google.com/patent/US10113774B2"
                  target="_blank" moz-do-not-send="true">https://patents.google.com/patent/US10113774B2</a><br>
              </p>
              <pre class="gmail-m_-4339801130891775252moz-signature" cols="72">Prentice</pre>
              <div class="gmail-m_-4339801130891775252moz-cite-prefix">On
                1/25/19 2:26 PM, Eric Moore wrote:<br>
              </div>
              <blockquote type="cite">
                <div dir="ltr">
                  <div dir="ltr">Actually, it looks like Joule-Thompson
                    cooling to me (Especially given the "Joule Force"
                    name). You've got the air intake (ambient), then an
                    expansion nozzle, into a low-pressure region, which
                    is created by the fan at the end. So the outlet
                    velocity of the air (and thus it's kinetic energy)
                    is higher than the inlet velocity, which would lower
                    the internal energy, and thus the temperature.
                    Instead the fins/nozzle/heatsink transfer heat to
                    the expanding gas, which exits a little above
                    ambient temperature. I imagine the drawback is you
                    really need to get rid of that high velocity hot
                    air, and can't recirculate it, or the kinetic energy
                    would be converted back to thermal energy, and mess
                    it all up. The descriptions do all involve the
                    exhaust air being ducted to the outside. This
                    article has the most technical detail: <a
href="https://www.nextplatform.com/2018/12/04/the-leading-edge-of-air-cooled-servers-leads-to-the-edge/"
                      target="_blank" moz-do-not-send="true">https://www.nextplatform.com/2018/12/04/the-leading-edge-of-air-cooled-servers-leads-to-the-edge/</a></div>
                </div>
                <br>
                <div class="gmail_quote">
                  <div class="gmail-m_-4339801130891775252gmail_attr"
                    dir="ltr">On Fri, Jan 25, 2019 at 11:33 AM Prentice
                    Bisbal via Beowulf <<a
                      href="mailto:beowulf@beowulf.org" target="_blank"
                      moz-do-not-send="true">beowulf@beowulf.org</a>>
                    wrote:<br>
                  </div>
                  <blockquote class="gmail_quote" style="margin:0px 0px
                    0px
0.8ex;padding-left:1ex;border-left-color:rgb(204,204,204);border-left-width:1px;border-left-style:solid">
                    <div bgcolor="#FFFFFF">
                      <p>You all know how much I like talking about heat
                        transfer and server cooling, so I decided to do
                        some research on this product:</p>
                      <p>Here's their website: <br>
                      </p>
                      <p><a
class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-txt-link-freetext"
                          href="https://forcedphysics.com"
                          target="_blank" moz-do-not-send="true">https://forcedphysics.com</a><br>
                      </p>
                      <p>and here's their YouTube channel with 5 videos:<br>
                      </p>
                      <p><a
class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-txt-link-freetext"
href="https://www.youtube.com/channel/UClwWeahYGuNl0THWVz1Hyow/videos"
                          target="_blank" moz-do-not-send="true">https://www.youtube.com/channel/UClwWeahYGuNl0THWVz1Hyow/videos</a> </p>
                      <p>This is really nothing more than an air-cooled
                        heatsink. I'm afraid I'm going to have to call
                        BS on this technology for the following reasons:
                        <br>
                      </p>
                      <p>1. It still uses air as the primary cooling
                        medium. I just don't think air has adequate
                        thermal conductivity or thermal capacity to
                        serve modern processor, no matter what you do to
                        it. <br>
                      </p>
                      <p>2. In the videos, they present highly idealized
                        tests with no control to use for comparison. How
                        do I know I wouldn't get the same results doing
                        the same experiment but using a similar duct
                        fashioned out of sheet metal. <br>
                      </p>
                      <p>3. Using this technology means a complete
                        redesign of your server hardware and possibly
                        your racks.</p>
                      <p>4. None of the information in the videos or on
                        their website really explains how this
                        technology works, and what really differentiates
                        it from any other air-cooled heat sink. Most
                        people with a good invention are usually excited
                        to tell you how it works. Since they brag about
                        30 international patents for this, there's no
                        need to try to protect a trade secret. </p>
                      <p>5. This statement:</p>
                      <p> </p>
                      <blockquote type="cite">The fins work like teeth
                        in a comb, neatly orienting air molecules to
                        point in the same direction and arranging them
                        into columns. </blockquote>
                      <p>Based on my education, this statement seems to
                        be completely devoid of science. <br>
                      </p>
                      <p>This statement seems to defy the laws of
                        physics. Last time I checked, unless an atom or
                        molecule is at absolute zero, it has movement,
                        whether it's spinning or vibrating, or both, so
                        how can they get air molecules to line up all in
                        neat little rows, where the molecules are all
                        pointing the same way? </p>
                      <p>This also implies very laminar flow.  As fluid
                        velocity increases that the diameter of the
                        channel decreases, the Reynolds Number
                        increases. As the Reynold's number goes up,
                        turbulence increases, so mathematically, I would
                        expect this flow to be tubulent, and not
                        laminar. From my classes on heat transfer,
                        turbulent flow around the heat transfer surface
                        increases heat transfer, so laminar flow in this
                        case wouldn't be a good thing. <br>
                      </p>
                      <p>Until they can provide better comparisons with
                        real servers in real data center environments,
                        I'm going to classify this as "snake oil"<br>
                      </p>
                      <p><a
class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-txt-link-freetext"
                          href="https://en.wikipedia.org/wiki/Snake_oil"
                          target="_blank" moz-do-not-send="true">https://en.wikipedia.org/wiki/Snake_oil</a><br>
                      </p>
                      <pre class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-signature" cols="72">Prentice</pre>
                      <div
class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-cite-prefix">On
                        1/24/19 3:54 PM, <a
class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-txt-link-abbreviated"
                          href="mailto:Chuck_Petras@selinc.com"
                          target="_blank" moz-do-not-send="true">Chuck_Petras@selinc.com</a>
                        wrote:<br>
                      </div>
                      <blockquote type="cite"> <font size="2"
                          face="sans-serif">Well, this is interesting.</font>
                        <br>
                        <br>
                        <font size="2" face="sans-serif">"According to
                          Forced Physics’ <</font><a
href="https://urldefense.proofpoint.com/v2/url?u=https-3A__forcedphysics.com_&d=DwMFAw&c=-_uRSsrpJskZgEkGwdW-sXvhn_FXVaEGsm0EI46qilk&r=fawF3TRTwCqlaBkoLcxYCr4F4NRwCc64hmEgi9rHPpE&m=zr6lAlVphGxOQTXSElww9hGpqb9IZPik0_MN2v8Fqjs&s=lb4Hi9X8NKIYWe_e1RU3Cw4gr9Uz_B7n5pnCNY0ss3U&e="
                          target="_blank" moz-do-not-send="true"><font
                            color="blue" size="2" face="sans-serif">https://forcedphysics.com/</font>
                          [forcedphysics.com]</a><font size="2"
                          face="sans-serif">> chief technology
                          officer, David Binger, the company’s conductor
                          can help a typical data center eliminate its
                          need for water or refrigerants and shrink its
                          22-MW load by 7.72 MW, which translates to an
                          annual reduction of 67.6 million kWh. That
                          data center could also save a total of US $45
                          million a year on infrastructure, operating,
                          and energy costs with the new system,
                          according to Binger. “We are solving the
                          problem that electrons create,” he said."</font>
                        <br>
                        <br>
                        <font size="2" face="sans-serif">A Cooler Cloud:
                          A Clever Conduit Cuts Data Centers’ Cooling
                          Needs by 90 Percent</font> <br>
                        <a
href="https://urldefense.proofpoint.com/v2/url?u=https-3A__spectrum.ieee.org_energy_environment_a-2Dcooler-2Dcloud-2Da-2Dclever-2Dconduit-2Dcuts-2Ddata-2Dcenters-2Dcooling-2Dneeds-2Dby-2D90-2Dpercent&d=DwMFAw&c=-_uRSsrpJskZgEkGwdW-sXvhn_FXVaEGsm0EI46qilk&r=fawF3TRTwCqlaBkoLcxYCr4F4NRwCc64hmEgi9rHPpE&m=zr6lAlVphGxOQTXSElww9hGpqb9IZPik0_MN2v8Fqjs&s=VuDTSuinKPMpF6NCztFZkSGOVo3LD7MLjroIj_sn0ao&e="
                          target="_blank" moz-do-not-send="true"><font
                            color="blue" size="2" face="sans-serif">https://spectrum.ieee.org/energy/environment/a-cooler-cloud-a-clever-conduit-cuts-data-centers-cooling-needs-by-90-percent</font>
                          [spectrum.ieee.org]</a> <br>
                        <font size="2" face="sans-serif"><br>
                          <br>
                          Chuck Petras, PE**<br>
                          Schweitzer Engineering Laboratories, Inc<br>
                          Pullman, WA  99163  USA<br>
                        </font><a href="http://www.selinc.com/"
                          target="_blank" moz-do-not-send="true"><font
                            size="2" face="sans-serif">http://www.selinc.com</font></a><font
                          size="2" face="sans-serif"><br>
                          <br>
                          SEL Synchrophasors - A New View of the Power
                          System <</font><a
                          href="http://synchrophasor.selinc.com/"
                          target="_blank" moz-do-not-send="true"><font
                            size="2" face="sans-serif">http://synchrophasor.selinc.com</font></a><font
                          size="2" face="sans-serif">><br>
                          <br>
                          Making Electric Power Safer, More Reliable,
                          and More Economical (R)<br>
                          <br>
                          ** Registered in Oregon.<br>
                        </font> <br>
                        <fieldset
class="gmail-m_-4339801130891775252gmail-m_8157281066646160146mimeAttachmentHeader"></fieldset>
                        <pre class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-quote-pre">_______________________________________________
Beowulf mailing list, <a class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-txt-link-abbreviated" href="mailto:Beowulf@beowulf.org" target="_blank" moz-do-not-send="true">Beowulf@beowulf.org</a> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit <a class="gmail-m_-4339801130891775252gmail-m_8157281066646160146moz-txt-link-freetext" href="http://www.beowulf.org/mailman/listinfo/beowulf" target="_blank" moz-do-not-send="true">http://www.beowulf.org/mailman/listinfo/beowulf</a>
</pre>
                      </blockquote>
                    </div>
                    _______________________________________________<br>
                    Beowulf mailing list, <a
                      href="mailto:Beowulf@beowulf.org" target="_blank"
                      moz-do-not-send="true">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"
                      target="_blank" rel="noreferrer"
                      moz-do-not-send="true">http://www.beowulf.org/mailman/listinfo/beowulf</a><br>
                  </blockquote>
                </div>
              </blockquote>
            </blockquote>
          </div>
          _______________________________________________<br>
          Beowulf mailing list, <a href="mailto:Beowulf@beowulf.org"
            target="_blank" moz-do-not-send="true">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"
            target="_blank" rel="noreferrer" moz-do-not-send="true">http://www.beowulf.org/mailman/listinfo/beowulf</a><br>
        </blockquote>
      </div>
    </blockquote>
  </body>
</html>