(back to the list of all Panda meeting minutes)

Time

8 am PT

Attendees

Richard Dubois Wei Yang Fabio Hernandez Michelle Gower Peter Love Brian YannyJames Chiang Wen Guan Edward Karavakis Zhaoyu Yang Tim Jenness Jen Adelman-Mccarthy

Regrets


Agenda:

  1. Update
    1. bps panda and clustering 
    2. other bps panda submission
      1. Update from @zhaoyu, Panda was able to run massive (relatively:-) parallel very short jobs,  e.g. processing ~2600 quanta/task via ARC CE and via SLAC harvester  
    3. IDF to DF submission
    4. site issues
      1. do we still have Squid/NAT issue ?
        1. networking/squid/nat issue didn't show up in the tests mentioned by Zhaoyu above.
      2. status of the second ARC CE at USDF (not done year, interrupted by power outage and Wei's travel)
    5. panda installation at USDF
  2. Next steps
    1. status of replication and butler ingestion at DFs for the 2.2i/defaults/test-med-1 collections in /sdf/data/rubin/repo/dc2 repo. 

Notes:

  1. Panda was able to run massive (relatively:-) parallel very short jobs,  e.g. processing ~2600 quanta/task . They were all submitted by SLAC harvester, directly to SLURM. iDDS managed the submission in a way that running jobs from the previous tasks did not block jobs in the next task. 
    1. Data is not available in EU DFs so this is currently USDF only
  2. Though not seen in the above test, the scaling issue of USDF Squid and NAT still effect jobs (Logging to Google went through Squid while updating jobs status with Panda server at CERN went through NAT)
  3. We are now using both the roma and milano clusters/partitions
  4. Indigo IAM team responded to Wen's question about port 8443 vs 443. Believe they are working on a fix.
  5. Power issue stopped the progress of the Panda DB work until Friday
  6. Date replication from FrDF to UKDF (for DP0.2) continue. A subset of raw image for 2.2i/defaults/test-med-1 is already there. Working on butler ingestion.