<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=utf-8">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    Colleagues,<br>
    <br>
    Find the meeting minutes reproduced below and at<br>
    <br>
     
    <a class="moz-txt-link-freetext" href="https://halldweb.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_September_2,_2015#Minutes">https://halldweb.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_September_2,_2015#Minutes</a>
    .<br>
    <br>
      -- Mark<br>
      ________________________________________________<br>
    <br>
    <meta http-equiv="content-type" content="text/html; charset=utf-8">
    <div id="column-content" style="margin: 0px !important; color:
      rgb(0, 0, 0); font-family: 'Times New Roman'; font-size: medium;
      font-style: normal; font-variant: normal; font-weight: normal;
      letter-spacing: normal; line-height: normal; orphans: auto;
      text-align: start; text-indent: 0px; text-transform: none;
      white-space: normal; widows: 1; word-spacing: 0px;
      -webkit-text-stroke-width: 0px; background-color: rgb(255, 255,
      255);">
      <div id="content" class="mw-body" role="main" style="direction:
        ltr; color: black; border: none !important; padding: 1em;
        margin: 0px !important; background: white;">
        <h1 id="firstHeading" class="firstHeading" style="font-weight:
          bold;" lang="en"><span dir="auto">GlueX Offline Meeting,
            September 2, 2015</span>, Minutes</h1>
        <div id="bodyContent" class="mw-body-content"><span
            class="mw-headline" id="Minutes"></span>
          <div id="mw-content-text" dir="ltr" class="mw-content-ltr"
            lang="en">
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Present:</p>
            <ul style="list-style-type: square;">
              <li><b>CMU</b>: Curtis Meyer, Mike Staib</li>
              <li><b>FIU</b>: Mahmoud Kamel</li>
              <li><b>FSU</b>: Brad Cannon, Aristeidis Tsaris</li>
              <li><b>JLab</b>: Alex Barnes, Amber Boehnlein, Mark
                Dalton, Mark Ito (chair), Paul Mattione, Eric Pooser,
                Nathan Sparks, Justin Stevens, Simon Taylor, Beni
                Zihlmann</li>
              <li><b>UConn</b>: Richard Jones, James McIntyre</li>
            </ul>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">There is a<span
                class="Apple-converted-space"> </span><a rel="nofollow"
                class="external text"
                href="https://bluejeans.com/s/8xII/" style="color:
                inherit !important; padding: 0px !important;
                text-decoration: inherit !important; background:
                transparent;">recording of this meeting</a><span
                class="Apple-converted-space"> </span>on the BlueJeans
              site.</p>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Announcements">Announcements</span></h3>
            <ol>
              <li>The<span class="Apple-converted-space"> </span><b>location
                  of the nightly builds</b><span
                  class="Apple-converted-space"> </span>at JLab has
                changed.
                <ul style="list-style-type: square;">
                  <li>Old location example:
                    <ul style="list-style-type: square;">
                      <li><code>/u/scratch/gluex/nightly/2015-09-02/sim-recon</code></li>
                    </ul>
                  </li>
                  <li>New location example:
                    <ul style="list-style-type: square;">
                      <li><code>/u/scratch/gluex/nightly/2015-09-02/Linux_RHEL6-x86_64-gcc4.4.7/sim-recon</code></li>
                    </ul>
                  </li>
                </ul>
                <p style="widows: 3; orphans: 3; margin: 1em 0px;
                  line-height: 1.2em;">See the<span
                    class="Apple-converted-space"> </span><a
href="https://halldweb.jlab.org/wiki/index.php/Nightly_Builds_of_GlueX_Software"
                    title="Nightly Builds of GlueX Software"
                    style="color: rgb(85, 34, 0); padding: 0px
                    !important; text-decoration: underline; background:
                    transparent;">nightly builds wiki page</a><span
                    class="Apple-converted-space"> </span>for more on
                  the system.</p>
              </li>
              <li><b>Sim-recon 1.5.1 has been released</b>. The<span
                  class="Apple-converted-space"> </span><a
                  rel="nofollow" class="external text"
                  href="https://github.com/JeffersonLab/sim-recon/releases/tag/1.5.1"
                  style="color: inherit !important; padding: 0px
                  !important; text-decoration: inherit !important;
                  background: transparent;">release notes</a><span
                  class="Apple-converted-space"> </span>are posted at
                GitHub.</li>
              <li>We are now<span class="Apple-converted-space"> </span><b>deleting
                  CCDB SQLite databases older than one year</b>.
                Formerly we were never deleting the monthly files. See
                the<span class="Apple-converted-space"> </span><a
href="https://halldweb.jlab.org/wiki/index.php/SQLite-form_of_the_CCDB_database#Archive_of_SQLite_Files"
                  title="SQLite-form of the CCDB database" style="color:
                  rgb(85, 34, 0); padding: 0px !important;
                  text-decoration: underline; background: transparent;">wiki
                  page</a><span class="Apple-converted-space"> </span>describing
                the system for more information.</li>
              <li>After filling up last week, our<span
                  class="Apple-converted-space"> </span><b>work disk was
                  expanded</b><span class="Apple-converted-space"> </span>from
                14 to 15 TB. Further expansion will require movement of
                the partition to a new file server.</li>
              <li>Mailman.jlab.org has been blocked from off-site for
                the past week or two due to cyber attacks. This
                restricts access to the mail archives and list
                administration pages, but does not affect the operation
                of the email lists. Access is not restricted from inside
                the JLab firewall.</li>
              <li>Mark drew our attention to his<span
                  class="Apple-converted-space"> </span><a
                  rel="nofollow" class="external text"
href="http://argus.phys.uregina.ca/cgi-bin/public/DocDB/ShowDocument?docid=2793"
                  style="color: inherit !important; padding: 0px
                  !important; text-decoration: inherit !important;
                  background: transparent;">new note on GlueX builds</a>.</li>
              <li>Kei gave a report on the last launch. See<span
                  class="Apple-converted-space"> </span><a
                  rel="nofollow" class="external text"
href="https://halldweb.jlab.org/wiki/images/1/10/2015-09-02-offline_monitoring.pdf"
                  style="color: inherit !important; padding: 0px
                  !important; text-decoration: inherit !important;
                  background: transparent;">his slides</a><span
                  class="Apple-converted-space"> </span>for the details.
                <ul style="list-style-type: square;">
                  <li>There were a lot of problems this time.</li>
                  <li>Kei continues to feedback to Chris Larrieu on SWIF
                    developments.</li>
                  <li>The memory footprint of the jobs nearly doubled.</li>
                  <li>The next launch is scheduled for September 12.</li>
                </ul>
              </li>
            </ol>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline"
                id="Review_of_minutes_from_August_19">Review of minutes
                from August 19</span></h3>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">We went over<span
                class="Apple-converted-space"> </span><a
href="https://halldweb.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_August_19,_2015#Minutes"
                title="GlueX Offline Meeting, August 19, 2015"
                style="color: rgb(85, 34, 0); padding: 0px !important;
                text-decoration: underline; background: transparent;">the
                minutes</a>.</p>
            <h4 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Hall_D_Package_Manager">Hall D
                Package Manager</span></h4>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Nathan made changes to support the
              use of a version.xml to create templates, as per feedback
              from the meeting.</p>
            <h4 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Flash_ADC_Pulse_Processing">Flash
                ADC Pulse Processing</span></h4>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">David convened a meeting of the
              interested parties. We will ask for a report at a future
              meeting.</p>
            <h4 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline"
                id="Who_is_responsible_for_deleting_Git_branches.3F">Who
                is responsible for deleting Git branches?</span></h4>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Our policy of having the branch
              creator delete branches after they have been merged is
              still in place, though compliance has been spotty.</p>
            <h4 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Clang_on_the_CUE">Clang on the
                CUE</span></h4>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Nathan reported that Clang is already
              in stalled on the ifarm, but was compiled against GCC
              4.8.3, whereas the default compiler there is GCC 4.4.7.
              That means that all packages might have to be compiled
              with the newer version. The system for supporting that for
              GlueX software has not been worked out.</p>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Geant4_Update">Geant4 Update</span></h3>
            <ul style="list-style-type: square;">
              <li>Richard has now configured the geometry to have the
                same magnetic field in all layers. This makes it
                insensitive to the algorithm Geant4 uses for picking the
                field-setting geometry layer. He is deferring an
                evaluation of the CPU cost of keeping the field on in
                the calorimeter volumes for a later stage of development
                work.</li>
              <li>He looked at how the current implementation performs
                with multi-threading turned on in Geant4. Initially the
                program was crashing. He traced the problem to a bug in
                the way geometry information is shared by threads. David
                had previously seen this and a fix was attempted by the
                Geant4 team. Unfortunately, the fix leads to other
                serious problems. Richard has put together his own patch
                in user code so he can keep going. He has also
                communicated his ideas for a more fundamental fix to the
                Geant4 folks.</li>
              <li>With multi-threading working properly, he sees perfect
                scaling of performance on an 8-core machine, up to 6
                threads (as high as he tried).</li>
              <li>The next step is to put together an end-to-end scheme
                for hit generation for a single detector system. That
                way others who might want to contribute will have a
                complete example to follow. He is planning on doing this
                with the CDC.</li>
            </ul>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Collaboration_Meeting_Agenda">Collaboration
                Meeting Agenda</span></h3>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">We discussed the agenda for the
              upcoming<span class="Apple-converted-space"> </span><a
href="https://halldweb.jlab.org/wiki/index.php/GlueX-Collaboration-Oct-2015"
                title="GlueX-Collaboration-Oct-2015" style="color:
                rgb(85, 34, 0); padding: 0px !important;
                text-decoration: underline; background: transparent;">Collaboration
                Meeting</a>, October 8-10, 2015 at Jefferson Lab.
              Possible talks and speakers are:</p>
            <ul style="list-style-type: square;">
              <li>Overview talk in including data challenge 3 report
                (Mark)</li>
              <li>Geant4 Development Update (Richard)</li>
              <li>Version Management System (Mark)</li>
              <li>Offline Monitoring and SWIF (Kei)</li>
              <li>Calibration Overview (Sean Dobbs)</li>
            </ul>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Data_Challenge_3_update">Data
                Challenge 3 update</span></h3>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Mark has run through the fake raw
              data set twice now. He showed<span
                class="Apple-converted-space"> </span><a
                href="https://halldweb.jlab.org/wiki/index.php/Data_Challenge_3"
                title="Data Challenge 3" style="color: rgb(85, 34, 0);
                padding: 0px !important; text-decoration: underline;
                background: transparent;">plots</a><span
                class="Apple-converted-space"> </span>of the number of
              jobs in various stages of batch farm progress as a
              function of time. The fluctuations are large and depend
              largely on what other users on the farm are doing.</p>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">He is currently working on converting
              the job submission process to use SWIF to see if that
              helps throughput.</p>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Commissioning_Simulations">Commissioning
                Simulations</span></h3>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Mark transmitted some thoughts from
              Sean, sent in an email.</p>
            <ul style="list-style-type: square;">
              <li><b><a
href="https://halldweb.jlab.org/wiki/index.php/Spring_2015_Commissioning_Simulations"
                    title="Spring 2015 Commissioning Simulations"
                    style="color: rgb(85, 34, 0); padding: 0px
                    !important; text-decoration: underline; background:
                    transparent;">Spring 2015 Commissioning Simulations</a></b>.
                The plan is to generate a few thousand files with the
                latest tagged sim-recon version only for run 9306.</li>
              <li><b>Fall 2015 Commissioning Simulations</b>. Sean
                thought we should start a discussion about how to set
                conditions for simulations for a possible Fall run. The
                Calibration Group has already decided to require a
                separate CCDB variation name.
                <ul style="list-style-type: square;">
                  <li>Justin thought that we should have production
                    underway by the time of the collaboration meeting so
                    people can estimate yields for the run.</li>
                  <li>Given the uncertainty in the conditions we might
                    have, we thought that we will probably have to just
                    pick some parameters and simulate those, say a 9 GeV
                    coherent edge and 1300 A magnet current.</li>
                </ul>
              </li>
            </ul>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline"
                id="Some_problems_with_F125_algorithms.3F">Some problems
                with F125 algorithms?</span></h3>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">Mike led us through<span
                class="Apple-converted-space"> </span><a rel="nofollow"
                class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2015-August/002138.html"
                style="color: inherit !important; padding: 0px
                !important; text-decoration: inherit !important;
                background: transparent;">his recent email</a><span
                class="Apple-converted-space"> </span>describing
              problems related to FADC hit in the BCAL with negative
              pulse integrals. The problem occurs for Mode 7 data and
              can be understood by looking at Mode 8 data where the
              read-out threshold is greater than 1/2 the peak amplitude.
              The result is early times for pulses late in the window.</p>
            <p style="widows: 3; orphans: 3; margin: 1em 0px;
              line-height: 1.2em;">We noted that there are two
              approaches on could take for Mode 8 data: emulated the
              Mode 7 algorithm, bugs and all, or modify the algorithm to
              give rational results in the problematic cases. There is a
              group meeting (see Review of Minutes above) on the
              emulation and we thought that they would very likely be
              working on an formulating an approach to the problem.</p>
            <h3 style="page-break-after: avoid; font-weight: bold;"><span
                class="mw-headline" id="Action_Items">Action Items</span></h3>
            <ul style="list-style-type: square;">
              <li>Convert DC3 to use SWIF (Mark)</li>
              <li>Re-do Spring simulation with new release (Mark)</li>
              <li>Implement a complete hit generation scheme for the CDC
                in Geant4 (Richard)</li>
            </ul>
          </div>
          <div class="printfooter" style="padding: 1em 0px;">Retrieved
            from "<a dir="ltr"
href="https://halldweb.jlab.org/wiki/index.php?title=GlueX_Offline_Meeting,_September_2,_2015&oldid=69729"
              style="color: rgb(85, 34, 0); padding: 0px !important;
              text-decoration: underline; background: transparent;">https://halldweb.jlab.org/wiki/index.php?title=GlueX_Offline_Meeting,_September_2,_2015&oldid=69729</a>"</div>
        </div>
      </div>
    </div>
    <div id="footer" role="contentinfo" style="color: rgb(0, 0, 0);
      margin-top: 1em; border-top-width: 1px; border-top-style: solid;
      border-top-color: rgb(170, 170, 170); direction: ltr; font-family:
      'Times New Roman'; font-size: medium; font-style: normal;
      font-variant: normal; font-weight: normal; letter-spacing: normal;
      line-height: normal; orphans: auto; text-align: start;
      text-indent: 0px; text-transform: none; white-space: normal;
      widows: 1; word-spacing: 0px; -webkit-text-stroke-width: 0px;
      background: white;">
      <ul id="f-list" style="list-style-type: square;">
        <li id="lastmod">This page was last modified on 3 September
          2015, at 21:02.</li>
      </ul>
    </div>
    <br>
  </body>
</html>