EPSCI Group Meeting Nov. 16, 2020

The meeting time is 10:00am.

Connection Info:

You can connect using BlueJeans Video conferencing (ID: 253 300 597). (Click "Expand" to the right for details -->):

Meeting URL
 https://bluejeans.com/253300597?src=join_info

Meeting ID
253 300 597

Want to dial in from a phone?

Dial one of the following numbers:
+1.888.240.2560 (US Toll Free)
(see all numbers - https://www.bluejeans.com/premium-numbers)

Enter the meeting ID and passcode followed by #

Connecting from a room system?
Dial: bjn.vc or 199.48.152.152 and enter your meeting ID & passcode

Agenda:

1. Previous meeting
2. Announcements
   • Streaming Data Scientist position
   • SRO Workshop VII (zoom),(homepage) (indico)
   • SRO meeting cancelled this afternoon due to workshop
   • Fortnight paper for Mon. 11/23:

     Development of FPGA based phase alignment logic for the high speed protocol in HEP Experiments

3. Ongoing Activities
   • JLab Common Environment (CE) + SPACK (Thomas, Nathan)
   • EVIO-6 (Carl)
   • JANA2 (Nathan)
     • GlueX port
     • ACTS + EIC
   • A.I.
     • JLab AI webpage (Kishan)
     • Hydra (Kishan/Thomas)
     • A.I.I.
     • Experimental Controls (David)
   • CODA/DAQ Software Support
   • CLARA Support
   • SRO (Vardan)
     • Hall-B/D TriDAS
     • ERSAP
   • Offsite Computing
     • BigRed3 at IU (David)
     • NERSC (David)
     • OSG (Thomas)
4. GUI for Calorimeter calibration scripts (Hall-D Request) (Thomas)
5. Publications
   • Netherland eScience Center Guide
6. AOT

---

Minutes:

Attendees: David L., Carl T., Nathan B., Kishan R., Vardan G., Thomas B.

Previous Meeting

• SoLID
  • Carl reached out to Zhiwen who responded that they would be having a simulation meeting, but no general software meeting. No other info given.
  • DAQ system for SoLID will be different than others at JLab and will likely require our input. Carl will look into who is designing that.

Announcements

• Streaming Readout Workshop VII is happening this week. EPSCI SRO meeting for this afternoon cancelled

SPACK

• Nathan contacted Thomas, but not much movement yet
• David will see about helping.

EVIO-6

• Difficult part of documentation is done.
• Dave A. requested C-API be maintained as is.
  • Carl will look at this and make sure documentation and source is updated where needed. Not expected to be a significant amount of work.

JANA2

• GlueX Port progressing. Some issues with factory calls that seem to be unresolved, causing stop conditions. Working trough them.
• ACTS + EIC
  • Met with EIC group last week and Nathan updated on current work.
  • Path is clear on what is needed next.
  • Agreed to meet again in ~3 weeks

AI

• Kishan posted another page on SciComp to describe installing custom kernels for Jupyter. Still a few things to finalize, but mostly there.
  • Some new updates posted to JLab AI pages for review (David needs to approve)
• Hydra
  • Code jam week before last successful
  • The major refactoring of the hydra code is largely done and the system was running
  • Some issue with web page updating, but not considered a major problem and should be easily fixed
• Experimental Controls: Two positions entered into Peoplefluent but have not yet been approved

CLARA Support

• Some confusion on how conflicts with parameter names in global and local levels are handled. Some options discussed with Hall-B and it was decided the priorities in the existing code should be left.

SRO

• Indra lab test setup
  • Ben R. and Dave A. solved issue with Indra test setup where no hits were being seen in Vardan's testing the previous week
  • Currently VTP test setup provides 300MByte/sec/link data rate.
  • Able to get 1% deadtime fairly quickly with multiple threads. For 600MByte/sec aggregated (2 VTP streams) stream processing 2.5 cores are used for aggregation and 4 cores are used for hit finder.
  • 1% dead time that was also present during the beam test was due to the disruptor ring-buffer tuning. Problem solved, but expected to be optimized again for the large data volumes (will see).
  • Aggregate rate of ~600MB/s with no loss of the VTP frames was achieved.
• TriDAS
  • Sergey B. presented estimation of data rates at HAll-B SRO meeting last week.
  • Suggested the current design spec should be for 5GB/s streams with the ability to scale to multiple 5GB/s streams

Offsite

• Production running still at NERSC, PSC, IU. Documentation still needed for new NERSC and IU workflows so it can be handed over to Hall-D postdoc
• OSG sent first jobs into JLab
  • Jobs started but had issues running
  • May well have fully running jobs by end of week (depends on vacation schedule of key OSG personnel)

Publications

• Everyone was encouraged to add their most recent publications (including proceedings) to the EPSCI publications page so we can fill in the blank section of the home page.