<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">Hi Parveen<br>
      <br>
      There is implementation of arnoldi time-stepper for spectra
      calculation with parpack for nek5000. I think I've already sent
      you some info about it. The code you can find at<br>
      <a class="moz-txt-link-freetext"
        href="ftp://ftp.mech.kth.se/pub/adam/Nek5000/toolbox/">ftp://ftp.mech.kth.se/pub/adam/Nek5000/toolbox/</a><br>
      There are sfd and arnoldi set-ups based on ext_cyl example form
      repository. The restart of the stepper you can find in the code.
      We do it by setting istep to 0 (forces solver to go through
      BDF1->BDF2->BDF3->BDF3) and resetting some vectors to
      zero.<br>
      Best regards<br>
      Adam<br>
      <br>
      On 22/10/14 15:07, <a class="moz-txt-link-abbreviated" href="mailto:nek5000-users@lists.mcs.anl.gov">nek5000-users@lists.mcs.anl.gov</a> wrote:<br>
    </div>
    <blockquote
      cite="mid:mailman.8343.1413983245.3424.nek5000-users@lists.mcs.anl.gov"
      type="cite">
      <div dir="ltr">
        <div class="gmail_extra">Hello Paul</div>
        <div class="gmail_extra"><br>
        </div>
        <div class="gmail_extra">I dont want to restart in the middle.</div>
        <div class="gmail_extra"><br>
        </div>
        <div class="gmail_extra">Every time I want to start the time
          stepping from scratch. So I want to know how to reset
          everything for a fresh simulation. Every time ARPACK gives me
          a vector, I have to start with
          BDF1->BDF2->BDF3->BDF3... until final time. </div>
        <div class="gmail_extra"><br>
        </div>
        <div class="gmail_extra">I think accuracy of time stepper also
          matters since we are approximating exact evolution operator
          exp(A*t) with the time stepper.</div>
        <div class="gmail_extra"><br>
        </div>
        <div class="gmail_extra">I have read some papers on time
          stepping approach but the details are not clear.</div>
        <div class="gmail_extra"><br>
        </div>
        <div class="gmail_extra">Thanks</div>
        <div class="gmail_extra">praveen</div>
        <div class="gmail_extra"><br>
          <div class="gmail_quote">On Wed, Oct 22, 2014 at 5:06 PM, <span
              dir="ltr"><<a moz-do-not-send="true"
                href="mailto:nek5000-users@lists.mcs.anl.gov"
                target="_blank">nek5000-users@lists.mcs.anl.gov</a>></span>
            wrote:<br>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div style="overflow:hidden">Hi Praveen,<br>
                <br>
                I would guess you need to develop your own timestepper
                because<br>
                Nek uses BDFk & EXTk  to time march, where typically
                k=3.  At<br>
                start-up, it uses k=1 and k=2 to bootstrap.   If you
                start timestepping<br>
                in the middle of the simulation then you need to
                overwrite the<br>
                existing coefficients so that you can restart the
                timestepper,<br>
                plus you need to store your initial condition (which
                presumably<br>
                amounts to overwriting vx,vy,vz, and pr).<br>
                <br>
                There are others in the community who have done these
                kinds of<br>
                things (I have not), who might have some further
                comments.<br>
                <br>
                Best,<br>
                <br>
                Paul<br>
                <br>
                PS ----- Are you trying to find eigenvalues associated
                with<br>
                the spatial operators ?   i.e., does the accuracy of the<br>
                timestepper matter?  If not, you can likely get away
                with<br>
                k=1, in which case you don't need much history in the
                backward<br>
                difference formula or extrapolation.  Pressure, however,
                is<br>
                a bit of a tricky business because we use splitting and
                there<br>
                is a residual O(dt^k) impact...</div>
            </blockquote>
          </div>
          <br>
          <br>
        </div>
      </div>
      <br>
      <fieldset class="mimeAttachmentHeader"></fieldset>
      <br>
      <pre wrap="">_______________________________________________
Nek5000-users mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Nek5000-users@lists.mcs.anl.gov">Nek5000-users@lists.mcs.anl.gov</a>
<a class="moz-txt-link-freetext" href="https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users">https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users</a>
</pre>
    </blockquote>
    <br>
  </body>
</html>