Collaboration Internal Minutes of 132nd Steering Board, 22nd March 2023

Submitted by mverderi on

Present: Marc Verderi, Susanna Guatelli, Alberto Ribon, Lorenzo Pezzotti, Laurent Garnier, Soon Yung Jun, Makoto Asai (OB  liaison), Mihaly Novak,
Krzysztof Genser (in absence of Gunter Folger), Gabriele Cosmo, Vladimir Ivanchenko, Daren Sawkey, Alex Howard, Anna Zaborowska, Witek Pokorski,
Dmitri Konstantinov, Luciano Pandola, Dennis Wright, Ben Morgan, Shogo Okada, John Apostolakis (number: 21) 

Excused:


Introduction:

The main discussion item of this meeting was the discussion on the 2023 work plan.

1. Situation with respect to international difficulties

Nothing to report.

2. Regular items

2.1 Open requirements:

  • https://jira-geant4.kek.jp/secure/Dashboard.jspa?selectPageId=10000
  • 5303 on Alberto `GIDI - LEND Models : install new GIDI when ready and validate with updated LEND`
    • No news.
  • 5302 on Susanna `To have an extended example to retrieve directly from the simulation Auger electron energy and associated atomic transition`
    • Looking for a student to work on it.
  • 5301 on Vladimir I. `Model for positronium creation and annihilation`
    • In progress: Omrane Kadri plans to work on it this year.
  • 5203 on Vladimir U. "Improve the inelastic cross sections of anti_deuteron and anti_He3 at low energies (< 1-2 GeV/c).`
    • Waiting for further measurements by ALICE.
  • 5202 on Mihaly `Precision versus speed optimised EM physics simulation configuration for ATLAS`
    • This is to be considered more as a regular, continuous activity, rather than a requirement per se
    • To be closed.
  • 5201 on Marc `To extend "force collision" biasing to charged particles`
    • In the work plan.
  • 5010 on Alberto `Physics models for ions below 1 MeV/u for Boron Neutron Capture`
    • No news.
  • 5009 on Sebastien `Extend energy and material coverage of G4-DNA beyond DNA and liquid water`
    • In progress - but a long way to go.
  • 5008 on Sebastien `Make EPICS2017 models available in Geant4 (electrons, photons) as an alternative to Livermore`
    • Already implemented - should be closed.
  • 5007 on Vladimir G. `Improve electro-nuclear models`
    • On-going.
  • 5006 on Vladimir G. `Improve simulation of gamma induced neutron background`
    • Some progress made recently, but more validation is needed.
    • JLab is willing to contribute to the validation of gamma-nuclear.
  • 5002 on Makoto `Support for "sub-event" parallelism across G4 threads`
    • Phase-I developments planned for this year.

Note:
- Marc is collecting requirements from Rennes CM session in a Google doc.
- This doc will be soon circulated, and passed to people tasked per domain

2.2 Open SB actions:

  • https://jira-geant4.kek.jp/secure/Dashboard.jspa?selectPageId=10020)
  • 131.1 on Marc `To restart turn over on pub board members`
    • On-going, candidate contacted.
  • 129.2 on Marc `To discuss at the pub board level the possibility to have a system (eg JIRA) to track the discussion on new publications`
    • Pending.
  • 127.1 on Dmitri `geant-val and validation of EM physics and hadronic physics models`
    • Migrated some EM tests to geant-val.
    • Need to refine the approach.
    • Long work in progress.
  • 125.3 on Dennis `To contact the Livermore people about status of LEND usability/correctness`
    • They are working on GIDI, and Doug Wright is developing some processes and physics lists.
  • 125.1 on Makoto `To make available on CERN site the SLAC HN archive`
    • Briefly investigated after website migration, see last comment in https://gitlab.cern.ch/geant4/geant4-website/-/issues/64.
    • The bottom line is that it needs quite a bit of work to make this browsable/searchable.
    • John Apostolakis suggested setting up the site `offline` and then use a crawler to convert it to a static copy. Ben/John/Makoto will meet and discuss this, though it will be at a low priority and no guarantee of success.
  • 122.4 on Marc `To put requirements from CM 2021 -> 2022 in JIRA (updated title)`
    • In progress.
  • 122.1 on Ben `To address the issue of "initialization in parallel" (follow-up of 118.3-4)`
  • 120.6 on Marc `To renew our physics and software posters`
  • 120.5 on Gunter `To investigate possible solution(s) for bookkeeping number of downloads`
    • Website now has "Matomo" analytics: https://webeos.docs.cern.ch/matomo/
    • This records downloads, though not perfectly, and is a lower limit (e.g. won't record direct downloads from GitHub/GitLab).
    • Access is granted on a per-user basis, if you want to view the data, let Ben Morgan know for instructions on this.
  • 120.2 on Marc `To find back old technical notes`
    • No progress.
  • 114.2 on Koichi `Create Indico entries for every past publication.
    • Contact past Publication Board chairs to identify past publications.
    • Also consider using JIRA for tracking new publications`
    • JIRA project page was created
    • Publication members will be contacted by Koichi on this.
  • 100.2 on Makoto `to ensure about visibility of publication (94.4 follow-up)`
    • No progress, but related to the previous (Koichi) item.
  • 93.4 on Marc `To follow up bug report #1719 and create a statistical test for gflash.`
    • Recent discussion with Dmitri and Lorenzo on how to implement it.
  • 84.6 on Koichi `To communicate on the way to release G4Py`
    • In progress.
    • Makoto should be kept informed on this.
    • The idea is not to pythonize every public method; to find out the limit to this.
    • Converting UI commands to python could be enough.
    • Often users require python to avoid learning and writing C++.
  • 83.4-11 on Koichi `Toolkit Developers Guide, chapter 2 : on Intercoms coordinator`
    • No progress.
  • 82.1 on Ben `Study the feasibility of 1) merging granular libraries, 2) Dividing large physics library, 3) option to drop unnecessary modules from libraries.`

2.3 Physics performance issues

2.4 Computing performance issues

  • https://g4cpt.fnal.gov//g4p/admin/reportSB.html
  • 4 measurements for recent releases (11.1.ref-01, 11.1.ref-02, 11.1.1, 11.0.4) since the last SB meeting
    • no major/outstanding issues.
  • Thanks to Julia Yarba:successfully migrated to gcc 11.3.0 and updated related tools (Openspeedshop, igprof)
    • Observed a marginal (a couple of percents) CPU improvement (w.r.t 8.3.0).
    • Openspeedshop development and user support have ended.
  • Also Julia posted Geant4 CPU performance by the version (new PR plots since 10.5 with gcc 11.3):

2.5 Internal development issues

  • Issues stored on Gitlab.

2.6 Update on publications, conference presentations & proceedings

  • CHEP presentation from EM (Vladimir I., Daren)

2.7 Report from ECRC

  • Note : a report from ECRC is now part of the SB meetings regular items.
  • Presentation by Lorenzo.
  • A dedicated ECRC web page will be set-up.
  • Main message: CERN is in a good position to apply for grants.

2.8 Geant4 Collaboration Evolution TF

  • Report from TF
    • Slides from Anna.
    • Items that were agreed at Rennes but that are not yet implemented will be followed by e-mail discussion at least, and dedicated meeting if needed.
    • These items will be part of the next SB meeting agenda.
  • Creation of an Indico, internal to the Geant4 Collaboration ?
    • Suggested to Marc by Giovanni Santin.
    • With:
      • Indico entry for each SB meeting, with link to the web page of the agenda (restricted to SB members).
      • Indico page of the WG meetings open to the whole Collaboration.
      • Collaboration-internal minutes of Steering Board meetings to be available in the Indico page of those meetings.
      • Moreover, it would be used for internal seminars, etc.
  • Access to old internal minutes:
    • Since the migration to the new web site, the minutes under `geant4.web.cern.ch` can not be accessed anymore.
    • John Apostolakis reminds that `geant4-internal.web.cern.ch` instead of `geant4.web.cern.ch` will probably fix many broken links of this sort.

2.9 Open bug reports

https://bugzilla-geant4.kek.jp/show_bug.cgi?id=2504, laurent.garnier@irisa.fr, ASSIGNED, "HepRepXML functionality deprecated, HepRepFile not a suitable replacement"

3. 2023 Work plan

The work plan discussion starts with a round table of the main items in the WG, asking to underline if any item may generate conflict / require coordination among several WGs.

  • WG plans:
    • Advanced Examples
      • Maintenance and code review of existing examples
      • Three new examples:
        • Development of two new examples dedicated to the ATHENA mission (one in system testing)
        • Proton tomography (in system testing)
        • No conflicts
    • Documentation Management
      • Focus on December release with sporadic releases as required
        • (handling of small bug fixes needs to be addressed).
      • No conflicts with other Working Groups.
    • Electromagnetic Physics Processes
      • Focus on HEP application, HepEM, R&D activities (phys in crystal, MuElec), and consolidation of DNA.
      • No general migration anticipated this year.
    • Geometry and Transport
      • Simplification of touchable.
      • Review accuracy of boundary crossing for magnetic field transport, which potentially can change some results
      • VecGeom: development of surface-based modeler for GPU R&D; also simplification of design is being prototyped.
    • Hadronic Physics Processes
      • Expected INCL antiproton annihilation at rest.
      • Further improvements in the treatment of thermal neutrons.
      • Major problem in the energy resolution (too optimistic) for hadronic showers (seen in ATLAS TileCal and HEC).
    • Generic Processes and Materials
      • Biasing of charged particles is a major item.
      • Revision of GFlash shower parameterization.
      • Continuation of the work on Machine Learning approach to fast simulation.
      • No conflicts with other Working Groups.
    • Novice and Extended Examples
    • Particles and Tracking
      • Same work plan as last year.
    • Persistency
      • Regular maintenance; few development items.
    • Physics Lists and Validation Tools
      • Geant-val extension (i.e. more tests in it, expansion of the user base) and improvements.
      • Other items similar to last year.
      • There may be an overlap with the new task force which may need to be discussed.
    • Run, Event and Detector Responses
      • First phase of sub-event parallelism implementation and related example.
      • May need to reiterate with Geometry/Transportation for parallel world scoring.
    • Software Management
      • Modularization impacts all Working Groups, but needs to be discussed across the Collaboration.
        • Expected changes to libraries, but no code changes.
      • Reviewing location of tests under toolkit categories, migrating these under tests/ directory.
        • Again, a move rather than a change, and responsible people will be contacted beforehand.
      • Reviewing compiler flags which may trigger additional warnings in testing.
      • Ongoing evolution of CI infrastructure may overlap with Testing and QA, to be discussed based on their requirements.
    • Testing and Quality Assurance
      • Oversee testing shifts.
      • Release management and Q/A monitoring with Valgrind and Coverity.
      • Perform regular profiling and benchmarking of Geant4 releases.
      • Grid testing, Reproducibility tests.
      • A new item proposed by Eunju Jeon (IBS-Korea, Underground Physics Center) for testing Geant4 version for related physics programs.
    • User and Category Interfaces
      • More code fitting to modern c++ style.
      • Update python interface (cmake / new examples)
    • Visualization
      • Migration to Qt6.
      • Work on finishing VTK driver.
      • Refinements and extensions to the Open Inventor Qt Viewer.
      • iOS and Android Devices: develop visualization solutions for these devices.
      • tools_sg (TSG) driver based on g4tools (*)
        • (under development; offers rendering on native systems such as Apple/Metal).
  • Task Force plans:
    • R&D
      • Complete the surface-based approach.
      • AdePT:
        • validation of current prototype;
        • integration of surface-based geometry modeler;
        • non-uniform B-field;
        • experimental integration in experiment frameworks.
      • Celeritas:
        • Fast simulation.
        • Optical photon simulation on GPU (CaTS/Opticks integration).
    • Validation
      • Discuss with other WG and experiments.
      • Tutorial on how to use geant-val.
      • Collect feedback and discuss the plan of work.
  • General Work Plan
    • To be prepared by Gabriele this week.
    • General comment:
      • it would be nice to specify also how much overall effort by the Geant4 Collaboration is devoted to development and improvement and how much on maintenance and testing.
  • 59th Technical Forum on 6th of April
    • Dedicated to the plan of work for 2023.

4. LHCC Software Report : Geant4 and ROOT Foundations Update, with Experiment Views

  • Follow-up from November 2021 "Common Software Review Projects" review
    • Closed session, on 7th of March
    • Alberto attended: positive feedback from the reviewers - official report in preparation

5. CM in Hokkaido

  • News on Medical Tutorial preparation
    • Program has been discussed in a separate meeting
    • 3 days, in parallel of Collaboration Meeting
      • Will require care about the two programs
    • Technical Forum on the last day
      • Will the CM TF, following tradition of host-colored TF
      • Medical items will be the main ones
    • Wish has been expressed to have participation of young members as lecturers and help for hands-on sessions
  • Next points
    • Web page will open in April.
    • Registration will start early in May (hopefully on the week of May 8-12)

6. AOB

  • 6.1 Finishing up with the census
    • Ensure e-mail lists are all well updated.
  • 6.2 Events
    • Past events:
      • OB meeting on 1st of February
      • Geant4 tutorial course at Kyushu University, Japan, Dec/14-16, 2022.
      • Next events:
  • 6.3 New collaborators:
    • No new collaborators 
  • 6.4 Easy access to the License
  • 6.5 AOB
    • Feed-back from ~1.5 month production of new Jekyll-based website
      • Not easy to find the link to the code guidelines.
      • Raise a GitLab issue is the most efficient way to report problems.
      • In geant4.org the Download seems to be broken.
    • Demands from African countries:
      • Marc received several requests for tutorials in African countries
      • Seminar on March 3rd (Marc & Lydia Maigne (GATE)).
      • Ghana tutorial in Accra (Marc asked by Daniel Adjei, Paris-Saclay).
      • (re)Request from Sebastien Guembou (Cameroun).
      • Aside: Geant4 website shows ~700 unique visitors from Africa since the beginning of March.
    • Geant4 Space Users Workshop
      • In US, hopefully in the second half of 2023
        • Last meeting was in October 2019, at Xylokastro, Corinth province, Greece.
      • Two candidate locations
        • NASA/Langley + JLab
        • NASA/JPL

Meeting Date : Wednesday, March 22, 2023 - 22:00 CET

Meeting Date