Sync meeting 2024 03 12 - multixscale/meetings GitHub Wiki

MultiXscale WP1+WP5 sync meetings

Next meetings

  • Tue 9 April 2024 10:00 CEST
  • Tue 14 May 2024 10:00 CEST
    • clashes with ISC'24!
    • reschedule to Tue 7 May?

Agenda/notes 2024-03-12

attending:

  • Caspar van Leeuwen, Maxim Masterov, Xin An, Satish Kamath (SURF)
  • Alan O'Cais (CECAM, Univ. of Barcelona)
  • Kenneth Hoste, Lara Peeters (HPC-UGent)
  • Julián Morillo (BSC)
  • Pedro Santos Neves, Bob Dröge (UGroningen)
  • Neja Šamec (NIC)
  • Susana Hernandez, Elisabeth Ortega, Nadia Martins (HPCNow!)
  • Jean-Noël (Univ. of Stuttgart)
  • Thomas (Univ. of Bergen)

project planning overview: https://github.com/orgs/multixscale/projects/1

Project review 2024-02-19

  • cfr. post-mortem Zoom meeting on Fri 23 Feb 2024
    • meeting notes in shared drive (Project management material / Minutes / MultiXscale CoE review)
    • we need to be more clear how we're in scope of the call
  • awaiting official report of experts...
  • next project review after M30 (June'25)
    • report is due when?
    • in practice in Sept'25?

EuroHPC Summit 2024 (Antwerp, 18-21 March 2024)

  • https://www.eurohpcsummit.eu
  • Mon 18 Mar'24 (15:30-18:00): EuroHPC Demo Lab Vega (Kenneth)
  • Tue 19 Mar'24 (14:15-15:15): MultiXscale poster (Eli & co)
  • Tue 19 Mar'24 (11:30-13:00): 13-min talk in "Centres of Excellence: Deploying Flagship Codes on EuroHPC Supercomputers" session (Alan)
    • slides are due today?
  • Thu 21 Mar'24 (14:00-18:00): involvement in "Elevate and Collaborate: European HPC NCCs and CoEs Workshop" (Alan + Kenneth)
    • "Fishbowl" panel session (Kenneth)
    • Success stories (Alan)
      • slides due tomorrow (Wed 13 Mar)
  • If you hear something interesting during the summit, drop a message in the MultiXscale Slack

MultiXscale General Assembly (Antwerp, 21-22 March 2024)

  • slides to prepare for "Overview of planned work for second year - technical WPs" (Thu 21 Mar'24, 10:30-11:10)
    • Caspar/Satish & Kenneth/Lara as WP leads
    • call scheduled Wed 13 Mar 13:00 CET
  • WP6+WP7 - Overview of planned work for second year - training and dissemination WPs (Alan, Eli)
  • press release
    • try to get interview with Belgian press on this?
      • maybe Knack Data News?
    • Susana can prepare a press release

Notes

  • hackathon moments
    • focused time to work together on open PRs, tasks, etc.
    • in pairs?
    • half a day?
    • once or twice per month?
      • or just actively schedule a short call with prime suspect to review your PR?
    • specific (short) list of tasks that should get finished
    • Tue 26 Mar 08:30-11:00
    • Tue 16 Apr 08:30-11:00
    • plan next 2 meetings at every meetings
  • making pyMBE paper open access
    • ~$3,500 cost, can maybe come from MultiXscale (after amendment to grant)?
  • work by Davide on QuantumESPRESSO (QE)
    • starting to look into compiler tuning to close the gap between foss and intel installations of QE (like enabling -ffast-math, with test suite still passing)
  • Jean-Noël: I was just asking about CECAM because we could probably involve the waLBerla people. Someone in their team told me they could maybe contribute something on multiscale simulations. We'll have a Zoom call with the waLBerla team in the next 10 days to organise another CECAM school, and Rudolf could use that opportunity to test the waters w.r.t. involving them in the multiscale CECAM event.

Overview progress per WP

WP1 (Developing a Central Platform for Scientific Software on Emerging Exascale Technologies)

  • all MultiXscale key apps (ESPResSo, walBerla, LAMMPS, OpenFOAM, ALL) now available in software.eessi.io repo (issue #3)
  • initial steps on supporting RISC-V (Julian, Bob) - T1.4
    • EESSI layers
      • CernVM-FS client
      • compat layer for riscv64
      • software layer:
        • GCC
        • foss
        • GROMACS
    • new CernVM-FS repository: riscv.eessi.io (see also docs)
    • paper submission to RISC-V workshop at ISC'24
  • Test suite release v0.2.0
  • see also EESSI update meeting 7 Mar'24 (slides)
    • Sync server to support private Stratum 1s #177 & #157
    • Compat layer ReFrame tests on software.eessi.io instead of pilot.eessi-hpc.org #196
    • Add container for RiscV #200
    • Security updates openssl, zlib, glibc #197
    • Resolved LDAP/user ID issues #32, #199, #99
    • Plenty of software added (BWA, dask, GDAL, LAMMPS, matplotlib, PyQt5, OSU with CUDA, snakemake, waLBerla, WRF)
    • Disable smcuda btl in OpenMPI as workaround on neoverse_v1 #473
    • Check missing installations against develop to flag unmerged PRs #486 and #494, open
  • Deliverables this year: 1.3, "Report on stable, shared software stack" (UB) (due M24, end 2024)
  • D1.4 Support for emerging system architectures (M30, mid 2025)
  • D1.5 Portable test suite for shared software stack (M30, mid 2025)
  • D1.6 Advances of RISC-V Ecosystem (M48, end 2026)
  • D1.7 Report on consolidation of the shared software stack (M48, end 2026)

WP5 (Building, Supporting and Maintaining a Central Shared Stack of Optimized Scientific Software Installations)

  • bot release v0.4.0
  • ongoing: weekly support sync meetings
  • D5.3 "Report on testing provided software" (due M30, mid 2025)
  • D5.4 "Final report on supporting and maintaining the shared software stack" (due M48, end 2026)

WP6 (Community outreach, education, and training)

  • "Task 6.1 - Awareness raising in the community" (issue #83)
    • planning issue needs an update, since elevator pitch is there?
    • we need a shorter elevator pitch (~2min)
  • Deliverables this year: 6.2, "Training Activity Technical Support Infrastructure" (UiB) (due M24, end 2024)
  • D6.3 Interim report on Community outreach, WP6 Education, and Training (due M30, mid 2025)
  • D6.4 Final report on Community outreach, WP6 Education, & Training (due M48, end 2026)
  • see also https://www.multixscale.eu/multixscale-in-a-nutshell/
    • needs text (abstract)
    • better pictures for showcases
  • workshop early Dec'24 with NCC Slovenia "Slovenian Supercomputing Days"
    • about key apps
  • ambassador program
    • NCC Slovenia was asked to appoint a person to be trained as ambassador
      • what kind of profile needed?
    • NCC Sweden has a "train the trainers" session
    • we can do online tutorials where ambassadors can help, or be on-site helper for remote attendees together in a room
    • find ambassadors in Brussels (VUB), Luxembourg (MeluXina or Univ. of Luxembourg), etc.

WP7 (Dissemination, Exploitation & Communication)

  • updates to MultiXscale website
  • Neja is sending poster to Baltic NCC event
  • blog post for website
    • ~half a page (2 paragraphs) + include a graphic element
    • on use of EESSI in CI for pyMBE (Jean-Noël) + arXviz preprint
    • on EuroHPC Summit (before, planned activities + also after)
  • Deliverables this year: 7.2, "Intermediate report on Dissemination, Communication and Exploitation" (HPCNOW) (due M24, end 2024)
  • D7.3 Business and Sustainability Plan (due M42, mid 2026)
  • D7.4 Final report on Dissemination, Communication and Exploitation (due M48, end 2026)
  • D7.5 Scientific software stack provisioning (due M48, end 2026)

WP8 (Management and Coordination)

  • COLA CASTIEL2 is being signed (issue #70)
    • still missing for UiB, BSC, UGroningen
  • Deliverables this year: 8.5, "Project Data Management Plan - final" (NIC) (due M24, end 2024)
    • missing planning issue for this?

CASTIEL2

  • "CernVM-FS and EESSI" tutorial session for CASTIEL2 + EuroHPC Hosting Entities (Fri 15 Mar'24, Alan + Kenneth)

Notes of previous meetings

Template for sync meeting notes

TO COPY-PASTE

  • overview of MultiXscale planning
  • WP status updates
    • [SURF] WP1 Developing a Central Platform for Scientific Software on Emerging Exascale Technologies
      • [UGent] T1.1 Stable (EESSI) - due M12+M24
        • ...
      • [RUG] T1.2 Extending support (starts M9, due M30)
      • [SURF] T1.3 Test suite - due M12+M24
        • ...
      • [BSC] T1.4 RISC-V (starts M13)
      • [SURF] T1.5 Consolidation (starts M25)
    • [UGent] WP5 Building, Supporting and Maintaining a Central Shared Stack of Optimized Scientific Software Installations
      • [UGent] T5.1 Support portal - due M12
        • ...
      • [SURF] T5.2 Monitoring/testing (starts M9)
      • [UiB] T5.3 community contributions (bot) - due M12
        • ...
      • [UGent] T5.4 support/maintenance (starts M13)
    • [UB] WP6 Community outreach, education, and training
      • ...
    • [HPCNow] WP7 Dissemination, Exploitation & Communication
      • ...