EJFAT EPSCI Meeting May. 8, 2024

From epsciwiki
Jump to navigation Jump to search

The meeting time is 2:30pm.

Connection Info:


Agenda:

  1. Previous meeting
  2. Announcements:
  3. System Status:
  4. NERSC Test Development:
    1. Data Source:
      1. JLAB, CLAS12, pre-triggered events - 1 channel
    2. Data Sink:
      1. Perlmutter
      2. ERSAP
    3. Test Plans - JLab, ESnet, NERSC:
    4. 100 Gbps / 25 - 40 Node Test
  5. ORNL/ESnet/JLab IRI Testbed (similar to NERSC) - Ross Miller <rgmiller@ornl.gov> project code CSC 266
    1. Waiting on ORNL Internal Network Security
  6. ALS
    1. Socializing Experiment Agenda - In Review
    2. Hoping for First Runs Summer 2024
  7. JLab FEG/SRO - will use interim UDP solution for event sync
  8. EJFAT Reconfig meetings/Tools to explore options
  9. EJFAT Phase II
    1. Need long term solution for CP/Source event sync
    2. Implementation details in the DAOS gateway.
      1. Need to spec DAOS Use Cases ?
      2. Intel standing up special slack channel to discuss DAOS
      3. Connection Strategy to DAOS
      4. Specially when to keep track of how the FPGA would DMA event data cells in the future if it was a SmartNIC card. ( Cissie )
      5. daosfs01​ has 2 physical IB cards and can run 2 true engines with each CPU socket hosting one engine.
    3. Progress of multi FPGA and multi virtual LB control plane sw. ( Derek )
    4. Progress of FPGA architecture ( Peter and Jonathan )
      1. LB FW currently limited to 100 Gbps
      2. Reassembly work commencing soon
    5. Progress of finalizing a reassembly frame format (subordinate to 4.) ( Carl / Stacey )
    6. Progress on software development for NVIDIA Bluefield2 DPU data steering from NIC to GPU memory ( Amitoj/Cissie )
  10. Resources:
    1. HPDF
    2. EJFAT API
  11. AOT