<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=utf-8">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <p>Folks,</p>
    <p>Find the minutes below and at <a class="moz-txt-link-freetext" href="https://goo.gl/MsDKJA">https://goo.gl/MsDKJA</a> .</p>
    <p>  -- Mark</p>
    <p>_____________________<br>
    </p>
    <div id="globalWrapper">
      <div id="column-content">
        <div id="content" class="mw-body" role="main">
          <h2 id="firstHeading" class="firstHeading" lang="en"><span
              dir="auto">GlueX Offline Meeting, August 3, 2016, </span><span
              class="mw-headline" id="Minutes">Minutes</span></h2>
          <div id="bodyContent" class="mw-body-content">
            <div id="mw-content-text" dir="ltr" class="mw-content-ltr"
              lang="en">
              <p>Present:
              </p>
              <ul>
                <li> <b>CMU</b>: Naomi Jarvis, Curtis Meyer</li>
                <li> <b>IU</b>: Matt Shepherd</li>
                <li> <b>JLab</b>: Alexander Austregesilo, Amber
                  Boehnlein, Mark Ito (chair), Paul Mattione, Nathan
                  Sparks, Justin Stevens, Simon Taylor, Beni Zihlmann</li>
                <li> <b>MIT</b>: Maria Patsyuk, Cristiano Fanelli</li>
                <li> <b>NU</b>: Sean Dobbs</li>
                <li> <b>Regina</b>: Tegan Beattie</li>
              </ul>
              <p>You can <a rel="nofollow" class="external text"
                  href="https://bluejeans.com/s/a6NJ/">view a recording
                  of this meeting</a> on the BlueJeans site.
              </p>
              <h3><span class="mw-headline" id="Announcements">Announcements</span></h3>
              <ol>
                <li> <b>Database Server upgrade: 5.5.50-MariaDB</b>.
                  The <a rel="nofollow" class="external text"
                    href="https://halldweb.jlab.org/talks/2016/db_upgrade_2016-07-21.pdf">upgrade
                    was complete successfully</a> two weeks ago by Marty
                  Wise of CNI.</li>
                <li> <b>sim-recon-2.2.1</b>. This <a rel="nofollow"
                    class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2016-July/002427.html">sub-minor
                    release</a> fixes the double-free bug for the 2.2
                  branch. This is now being used by sim1.1.</li>
                <li> <b>Analysis TTree: Unused Tracks/Showers No Longer
                    Saved</b>. By default, <a rel="nofollow"
                    class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2016-July/002425.html">unused
                    hypotheses are not longer saved</a> in particle
                  combos. The old behavior can be recovered by setting
                  an argument to the tree-output-enabling function. This
                  results in a 40% reduction in disk space for the
                  recent Analysis Launch.</li>
              </ol>
              <h3><span class="mw-headline"
                  id="Review_of_minutes_from_the_last_meeting">Review of
                  minutes from the last meeting</span></h3>
              <p>We went over the <a
href="https://halldweb.jlab.org/wiki/index.php/GlueX_Offline_Meeting,_July_20,_2016#Minutes"
                  title="GlueX Offline Meeting, July 20, 2016">minutes
                  from the meeting on July 20</a>.
              </p>
              <ul>
                <li> <b>Managing Plugins</b>. We agreed that, at least
                  in the short term, we should pursue options 2 and 3
                  from the list:
                  <ul>
                    <li> 2. Reduce the number of plugins that are built
                      automatically. (Justin)</li>
                    <li> 3. Have finer-grained build targets for SCons
                      so that parts of the build can be skipped
                      depending on the needs of the user. (Sean)</li>
                  </ul>
                </li>
                <li> <b>Bad files in Lustre</b>. Mark and Alex have
                  reported a few more bad files resulting in lost data.
                  Alex has seen files go bad since the Lustre upgrade.</li>
              </ul>
              <h3><span class="mw-headline"
                  id="Report_from_SciComp_Meeting_on_July_21">Report
                  from SciComp Meeting on July 21</span></h3>
              <p>Mark reviewed the <a rel="nofollow" class="external
                  text"
                  href="https://halldweb.jlab.org/talks/2016/scicomp_2016-07-21.pdf">agenda
                  from the meeting</a>.
              </p>
              <ul>
                <li> We are getting 48 new farm nodes, with 18 cores
                  each, and 190 TB of disk.</li>
                <li> A software review is coming in November.</li>
                <li> There may be an overlap between a planned down of
                  the tape library and Hall D running.</li>
              </ul>
              <p>At this SciComp meeting Mark brought up the idea of
                charging for memory usage in the fairshare algorithm. It
                is a scarce resource; often times cores go idle because
                a farm node cannot accommodate more jobs due to memory.
                There was some discussion of how this could be done.
              </p>
              <p>With multi-threaded JANA jobs memory usage vis-a-vis
                CPU usage is efficient. Curtis remarked that we have put
                a lot of effort into optimizing resource usage. It was a
                huge commitment on our part and all groups should take
                that seriously across the Lab.
              </p>
              <h3><span class="mw-headline"
                  id="Report_from_Computing_Round_Table_on_August_2">Report
                  from Computing Round Table on August 2</span></h3>
              <p>Mark reported in the <a rel="nofollow" class="external
                  text"
                  href="https://halldweb.jlab.org/talks/2016/round_table_2016-08-02.pdf">meeting
                  held yesterday</a>. <a rel="nofollow" class="external
                  text" href="http://vis.lbl.gov/%7Ewes/">Wes Bethel</a>
                from LBNL described an ASCR workshop he organized that
                addressed data challenges faced by Office of Science
                experimental and observational science programs.
              </p>
              <p>In the morning before the meeting, Wes met separately
                with each of the Halls. We presented brief overviews of
                our computing model and mentioned "pain points" that we
                are feeling. Mark, Justin, Alex, and Sean participated
                in the Hall D session.
              </p>
              <p>Amber made some remarks:
              </p>
              <ul>
                <li> She solicited input on speakers and topics that
                  might be presented at these meetings. Four have been
                  held so far. Amber can often come up with a name from
                  outside the Lab to address a specific topic, and
                  speakers from inside the Lab are sought as well.</li>
                <li> <a rel="nofollow" class="external text"
                    href="https://www.jlab.org/indico/event/157">Wes's
                    slides are available</a> on the Lab's Indico site.</li>
                <li> Wes has noticed similarities in problems faced in
                  programs across the Office of Science programs, yet
                  the discussion at any one lab tends to emphasize
                  differences.</li>
                <li> She thinks that computing is a frontier area for
                  the 12 GeV program. In this context, specifically in
                  the areas of data management, computational science,
                  and visualization, we should look to learn from others
                  whenever possible.</li>
                <li> Mark will announce future meetings to the offline
                  group. Folks should feel free to participate when
                  topics stimulate personal interest.</li>
              </ul>
              <h3><span class="mw-headline" id="Analysis_Launch">Analysis
                  Launch</span></h3>
              <p>Alex described the Analysis Launch. It started last
                week and included contributions from 12 people. In all
                46 channels were analyzed. The jobs are finishing now.
                There was some competition from sim1.1 otherwise things
                would have gone faster. In all there are 4 TB of ROOT
                trees and another 1 TB of ROOT histograms. </p>
              <p>Alex led us through the Analysis Launch section (near
                bottom) of the <a rel="nofollow" class="external text"
href="https://halldweb.jlab.org/data_monitoring/launch_analysis/index.html">launch
                  analysis webpage</a>. The link there shows <a
                  rel="nofollow" class="external text"
href="https://halldweb.jlab.org/data_monitoring/analysis/summary_swif_output_analysis_2016-02_ver01_batch01.html">various
                  statistics and graphs of the jobs</a>.
              </p>
              <h3><span class="mw-headline" id="Next_Monitoring_Launch">Next
                  Monitoring Launch</span></h3>
              <p>Sean proposed that if the new tracking code is ready we
                should try another monitoring launch to test those and
                other improvements to the code. Simon thinks that the
                code is ready. We decided to try to do a launch this
                Friday.
              </p>
              <p>On a related note, folks thought that the time is right
                for Mark to do another sim-recon release.
              </p>
              <h3><span class="mw-headline" id="sim1.1">sim1.1</span></h3>
              <p>Mark reported on the [sim1_1 jproj status,
                8-3-16|status] of <a rel="nofollow" class="external
                  text"
                  href="https://halldweb.jlab.org/gluex_simulations/sim1.1/">sim1.1</a>.
                75% off the jobs and most of the resulting REST files
                have been written to tape.
              </p>
              <p>Justin suggested running an analysis launch on the
                sim1.1 REST files. We endorsed the idea, but did not
                find a volunteer.
              </p>
              <h3><span class="mw-headline" id="ROOT_6_Upgrade.3F">ROOT
                  6 Upgrade?</span></h3>
              <p>We discussed several points, but in the end agreed on
                the following course:
              </p>
              <ul>
                <li> For now we keep ROOT 5 as the default version.</li>
                <li> We maintain a parallel build of all packages
                  against ROOT 6. Early adopters can try this out.</li>
                <li> We run the next monitoring launch using the ROOT 6
                  version as a large scale test.</li>
                <li> The assumption is that all of our public compiled
                  code is compatible with both versions of ROOT. If that
                  turns out not to be true, then the scheme breaks down.</li>
                <li> Not all macros (they do not participate in the
                  build procedure) will be compatible with both versions
                  and so people running those will have to know what
                  they are doing. Since the default remains ROOT 5, old
                  macros will continue to work for the naive user.</li>
                <li> Once we have established that using ROOT 6 in the
                  monitoring launch is working, we will revisit the
                  issue.</li>
              </ul>
              <h3><span class="mw-headline"
                  id="Review_of_recent_pull_requests">Review of recent
                  pull requests</span></h3>
              <p>We reviewed the <a rel="nofollow" class="external
                  text"
href="https://github.com/JeffersonLab/sim-recon/pulls?q=is%3Apr+is%3Aclosed">list
                  of recent pull requests</a>.
              </p>
              <p>We mainly expressed shock and awe at the large number
                of <a rel="nofollow" class="external text"
                  href="https://github.com/JeffersonLab/sim-recon/pull/489">commits
                  in Simon's recent upgrade of the tracking code</a> to
                include timing information from the FDC anode wires.
                Corresponding changes to the CCDB have been entered.
              </p>
              <h3><span class="mw-headline"
                  id="Review_of_recent_discussion_on_the_GlueX_Software_Help_List">Review
                  of recent discussion on the GlueX Software Help List</span></h3>
              <p>We looked over <a rel="nofollow" class="external text"
href="https://groups.google.com/forum/#%21forum/gluex-software">recent
                  conversations</a>.
              </p>
              <p>Maria has got the <a rel="nofollow" class="external
                  text"
href="https://groups.google.com/forum/#%21topic/gluex-software/cGfeCq6G8aA">ROOT
                  visualization of the detector</a> going, though there
                are some features left to be desired. Simon mentioned
                that the hdgeant++ (CERNLIB) programs has tools that are
                useful for finding overlaps and the like.
              </p>
            </div>
            <div class="printfooter">
              Retrieved from "<a dir="ltr"
href="https://halldweb.jlab.org/wiki/index.php?title=GlueX_Offline_Meeting,_August_3,_2016&oldid=76654">https://halldweb.jlab.org/wiki/index.php?title=GlueX_Offline_Meeting,_August_3,_2016&oldid=76654</a>"</div>
          </div>
        </div>
      </div>
      <div id="footer" role="contentinfo">
        <ul id="f-list">
          <li id="lastmod"> This page was last modified on 4 August
            2016, at 11:20.</li>
        </ul>
      </div>
    </div>
    <pre class="moz-signature" cols="72">-- 
<a class="moz-txt-link-abbreviated" href="mailto:marki@jlab.org">marki@jlab.org</a>, (757)269-5295
</pre>
  </body>
</html>