GlueX Software Meeting, April 7, 2025
From GlueXWiki
GlueX Software Meeting
Monday, April 7, 2025
11:00 am EDT
F326/327
Zoom Meeting ID: 160 636 9159 Passcode: 888788 Join
Mark Ito is inviting you to a scheduled ZoomGov meeting.
Topic: GlueX Software Time: This is a recurring meeting Meet anytime
Join ZoomGov Meeting https://jlab-org.zoomgov.com/j/1606369159?pwd=SlBrdStCQzllano1SmVQazMwaFExdz09
Meeting ID: 160 636 9159 Passcode: 888788 One tap mobile +16692545252,,1618692159# US (San Jose) +16468287666,,1618692159# US (New York)
Dial by your location
+1 669 254 5252 US (San Jose) +1 646 828 7666 US (New York) +1 669 216 1590 US (San Jose) +1 551 285 1373 US 833 568 8864 US Toll-free
Meeting ID: 160 636 9159 Find your local number: https://jlab-org.zoomgov.com/u/acAwo1X4w9
Join by SIP 1618692159@sip.zoomgov.com
Join by H.323 161.199.138.10 (US West) 161.199.136.10 (US East) Meeting ID: 160 636 9159 Passcode: 888788
Agenda
- Announcements
- Duplicated files will be deleted in ~50 days ,check yours with jcache duplicatedFileInfo halld (-l)
- GlueX Tutorial 2025 : May 12
- New cooking skim proposed by Nathan is working. Use 5 32 threads jobs on one node instead of running 1 32 threads job on one node. => Cost reduction by a factor 4.5 and will consume ~50k node hours for 2023-01 data set
- 6 16 threads jobs on one node, 43-perl, id=37489380, rate=146.79Hz, success efficiency=.9646, cost 101.23 node hrs, processing time 11043s
- 5 32 threads jobs on one node, 42-perl, id=37480365, rate=188.95Hz, success efficiency=.9494, cost 96.35 node hrs, processing time 8894s
- 5 32 threads jobs on one node, 45-perl, id=37498699, rate=189.09Hz, success efficiency=.9494, cost 95.63 node hrs, processing time 8827s
- 4 64 threads jobs on one node, 44-perl, id=37498697, rate=173.53Hz, success efficiency=.9545, cost 131.46 node hrs, processing time 9658s
- Review of minutes and action items
- Software status:
- Pull_request test currently running on halld-cron-2024 (RHEL9), but connection from halldweb refused (Wesley)
- Container status
- GitLab
- MCWrapper status:
- Project to switch containers in-flight: #134
- RUCIO update (Anil)
- Update on new halld_recon libraries:
- Helicity
- TRD
- ECAL
- Discussion of software upgrade projects:
- JANA2 (Raiqa)
- Transition to JANA2 wiki page
- Locking, CCDB access and optimal thread counts
- RCDB (Dmitry)
- Upgraded database: gluondb1.jlab.org/rcdb2 or hallddb.jlab.org/rcdb2, automatically selected by gxenv
- DAQ is writing to rcdb2
- New (temporary) website: rcdb.pythonanywhere.com/runs/, moving to halldwebdev (RHEL9, Sherman)
- A (desperate) sanity saving workaround for no-gui days
- ROOT
- version 6.32.08 in counting house, compiled with c++20
- minor changes needed for c++20 in hdds, amptools
- JANA2 (Raiqa)
- Review of recent issues and pull requests:
- Review of recent discussion on the GlueX Software Help List (all)
Notes
Action Items
- Documentation
- Improve documentation on singularity containers
- Check if native build works for container on Alma9, eventually get rid of -cntr builds
- Software Upgrades
- halld_recon:
- $HALLD_RECON_HOME/src/BMS is deprecated, remove from the repo?
- Remove dependency on histogram tools [1] (Sean, partly done)
- #842: Locking in JANA, CDC plugins fixed
- Make large 2D histograms in monitoring_hists optional: request for OSG production
- Implement decoding and monitoring for TRD prototype (Sean, Simon, Lubomir, Nizar)
- Implement decoding and monitoring for beam helicity measurements (Mark D. Justin, Ken, Peter)
- gluex_MCWrapper:
- MCWrapper needs possibility to switch between operating systems / containers during workflow: #134 created
- RCDB (Dmitry):
- Fix bug in upgrade command (Sean, Dmitry)
- Move new webpage to halldweb (Dmitry, Sherman)
- halld_recon:
- Broken tape
- IBM was not able to recover, data lost
- 14839 files (1357 REST) from 2017-01 ver04, about 5%
- Full list, reprocess?