EJFAT EPSCI Meeting Aug. 21, 2024

From epsciwiki
Jump to navigation Jump to search

The meeting time is 2:30pm.

Connection Info:


Agenda:

  1. Previous meeting
  2. Announcements:
    1. ejfat-1 LB up and should behave the same as ESnet Stable LB
    2. ejfat-2 LB up and should behave the same as ESnet Stable LB
  3. Topics
    1. IRI Test Development:
      1. LB version = ESnet Stable version
      2. Data Source:
        1. JLAB, CLAS12, pre-triggered events - 1 channel
      3. Data Sink:
        1. Perlmutter - 40 nodes
        2. ORNL/ESnet/JLab IRI Testbed / Defiant - 4 nodes allocated
        3. JLab - 7 nodes available
        4. ERSAP
      4. Test Plans - JLab, ESnet, NERSC:
      5. Prometheus Dashboards
    2. JLab FEG/SRO
      1. will use interim UDP solution for event sync
      2. Sparse Port Map Issue
      3. Network Fabric Entropy
    3. E2SAR
      1. e2sar/ibaldin:0.1.0a6 available
      2. segmentation done, reassembly not completed yet
      3. Port Map Issue
    4. IB
    5. ejfat-3- should't be used until further notice
      1. needs configuration corrections for networking, routing, mounting /daqfs
      2. Hosts 2 U280 FPGAs
    6. ACAT 2024 Paper can fold in improvements with reviewer comments
    7. Awaiting TLS cert install across cluster
    8. Turn machines over to sys admin/Puppet - except for ejfat-fs
    9. SC poster submitted - demo in works
    10. SSD drives on ejfat-fs - 20TB used of 28TB - mount for EJFAT farm access probably ZFS/Raid for data preservation
    11. Ram Disks: 1TB Total Mem on ejfat-fs, 0.5 TB others
    12. Lustre storage for EJFAT - not available for Testbed or Ubuntu 20.04
    13. Experiment Halls - beam returns late January
  4. Resources:
    1. HPDF
    2. EJFAT API
    3. EJFAT Status
  5. AOT