Login

WLCG Management Board #332

→ Europe/Zurich
513/R-068 (CERN)

513/R-068

CERN

19
Show room on map
Simone Campana (CERN)
Description

16:00 CERN/10:00 EDT/09:00 CDT

URL for the ZOOM call: https://cern.zoom.us/j/66011416092  (the passcode has been distributed by email)

To join by phone (see the dedicated support page):
1) dial your local ZOOM number (e.g. +41432107108) then enter the MB meeting ID  660 1141 6092 then #
2) mute your phone (there is no auto-mute facility)
Note: calls are charged to the caller (there is no callback facility)

Email Distribution List: worldwide-lcg-management-board@cern.ch
Email List Archive: worldwide-lcg-management-board (requires CERN authentication)
Minutes: Management Board Meeting Minutes
Action List: MbActionList
See also the: WLCG Document Repository,  WLCG Web Site

WLCG Management Board
Zoom Meeting ID
66011416092
Host
Andrea Sciabà
Alternative hosts
Simone Campana, Maarten Litmaath
Useful links
Join via phone
Zoom URL
    • 16:00 → 16:20
      Minutes and Matters Arising 20m
      • Resource model for Run-4: call for contributors 5m
        Speaker: Simone Campana (CERN)

        As discussed at the last Collaboration Board meeting we should form a small group of volunteers to build a first proposal of the  WLCG resource model for Run-4. The process should be iterative and the CB/MB should be informed at regular intervals. The goal is to have a first proposal in place for the June CB

         

      • WLCG workshop 5m
        Speaker: Simone Campana (CERN)

        Given the CHEP schedule in 2026 and the WLCG workshop on heterogeneous architectures in Dec 2025 (see next point) we agreed to shift the WLCG workshop from May 2026 to September/October 2026. We are still looking for a location and we welcome proposals. 

      • Heterogeneous Architectures Workshop 5m
        Speaker: Simone Campana (CERN)

        The WLCG workshop on Heterogeneous architectures will take place on Dec 3-5 2025. Contrarily to the initial plans (2 days) we will have a 2.5 days workshop, starting in the Morning of Wednesday the 3rd do December and finishing at lunchtime of Friday the 5th of December 

        https://indico.cern.ch/event/1526077/overview

         

        Initial thoughts about the structure of the workshop: 

         

        WLCG GPU workshop in December

        Logistics: Wednesday morning to Friday morning (3–5 Dec). Dinner (TBD) on either Wednesday or Thursday evening (Luigia? Fondue?).

        Sessions:

        • Progress of experiments and projects (e.g. Geant4) with the porting of classic applications on GPUs (generators, simulation, reconstruction), including expected roadmaps to production.

        • ML/AI activities that require GPUs for training and possibly inference.

        • Current offering from sites and discussion of requirements for future provisioning.

        • Cross-cutting discussion on “what problem are we trying to solve”, focusing on key metrics (provisioning models, CPU–GPU balance, offload factor, memory/I/O needs, scalability and portability).

        • Final session to summarize and define the work plan and next steps.

        Organization: leading team of ~4 people from TCB (Ale, James), software liaisons (Stefan, Predrag), and one person from the ML community (e.g. IML, EuCAIF, ML4HEP), with delegated tasks for setting up the scientific agenda.

        Example Goals:

        • Understand the current status of the code and workflows that can run on GPUs - identify commonalities and differences.
        • Understand what fraction of production computing can currently run on GPUs and how that fraction might evolve between now and Run-4.
        • Start to discuss if and when pledges for GPU resources could be accepted or required.
        • (Re)discuss the needs of the experiments and how to evolve from today’s small-scale use to scalable, production-level integration in WLCG.
        • Agree on(discuss?) the next steps towards a  workplan for benchmarking, provisioning, and integration of GPUs into WLCG.

         

      • Support letter for Horizon call for NeIC 5m
        Speaker: Simone Campana (CERN)

        From the NeIC director (not verbatim):

        ##

        NeIC is applying for Horizon call: [HORIZON-INFRA-2025-01-DEV-01]. This call requires:

        "proposals should include at least one ESFRI Landmark or European Research Infrastructure Consortium (ERIC) as beneficiary. In case of a distributed ERIC, as an alternative to the ERIC participating as a beneficiary, a legal entity that is hosting ERIC facilities, resources or related services may participate as a beneficiary."

         

        The essence of what we apply for is to support technical staff at research infrastructures as follows:

        developing a comprehensive and accessible training model, facilitating training in response to the identified specific needs of technical staff and researchers , in particular Cybersecurity, FAIR research software and data stewardship. The work will build on existing frameworks, such as e.g. used within CodeRefinery, which have already proven to be role models in the cross-border training activities.

         

        WLCG is a natural candidate for seeking support due to our history with NT1. The support letter will not be binding for WLCG in any way - if it goes through, we'll be good to discuss further how WLCG could benefit from it.

         

         

        ##

    • 16:20 → 16:25
      Action List Review 5m
    • 16:25 → 16:35
      WLCG Service Report 10m
      Speakers: Panos Paparrigopoulos (CERN), Maarten Litmaath (CERN)
      WLCG-svc-report-250916-v10.pdf
      WLCG-svc-report-250916-v10.pptx
    • 16:35 → 16:45
      TCB Report 10m
      Speakers: Alessandro Di Girolamo (CERN), James Letts (Univ. of California San Diego (US))

      Since the previous WLCG MB, we organized a meeting of the WLCG Technical Coordination Board (TCB#6, Indico: https://indico.cern.ch/event/1568456/) on Thursday, August 26th to checkpoint progress and plans on the various chapters of the WLCG Technical Roadmap. Most of the team is in place with a few exceptions, mostly that need to be negotiated with working or development groups outside of the WLCG such as for FTS and benchmarking. The next TCB meeting will likely be near the end of the month.

      Note that there were no questions for the TCB specifically during the quarterly meeting with the WLCG LHCC Referees on Tuesday, September 2nd, Indico (restricted): https://indico.cern.ch/event/1525773/. 

      We also organized a meeting of the  WLCG Open Technical Forum on Tuesday and Wednesday, September 9th-10th (OTF#6, Indico: https://indico.cern.ch/event/1562124/). The agenda concentrated on environmental sustainability on Tuesday and DOMA-related activities (recent data mini-challenges, IPv6 activities and plans) on Wednesday. Thank you to the conveners of the WLCG Sustainability Forum and the DOMA conveners for their efforts to organize the sessions. In the environmental sustainability session there was an interesting presentation about MadGraph improvements using vectorized instructions on CPUs and GPU. In the Wednesday session we learned about progress on IPv6 plans and many interesting lessons learned from the recent mini data challenges. It could be very useful to perform such challenges in other regions.

      Regarding the WLCG action item #230620-3 on GPU benchmarking, note that the Technical Coordinators met with the main HEPScore developers on Tuesday, September 2nd to better understand their situation and the next steps. 

      A short report for the WLCG Report for the October RRB has been sent to the LCG Office.

    • 16:45 → 16:55
      WLCG Token Profiles v1.1 10m
      Speaker: Maarten Litmaath (CERN)
      profile-vXY.pdf
      WLCG-Token-Profiles-v1.1.pdf
      WLCG-Token-Profiles-v1.1.pptx
    • 16:55 → 17:00
      AOB 5m
      • Next MB Meeting: Tuesday 14 October 2025 1m
Powered by Indico v3.3.9-pre