---

HSF Weekly Meeting #128, 1 March, 2018

Present/Contributors: Graeme Stewart, Mohammad Al-Turany, Torre Weanus, Michel Jouvin, Riccardo Maria Bianchi, Helge Meinhard, Jim Pivarski, David Lange, David Crooks, John Harvey, Ian Collier, Martin Ritter, Ricardo Maria Bianchi, Sudhir Malik, Eduardo Rodrigues, Dario Menasce, Mark Neubauer, Martin Ritter, Oli Gutsche, Ian Collier, Simone Campana

HSF/WLCG Workshop

• General points

  • 175 participants registered now.
  • Extended early registration extended up to and including Monday (5 March) after some confusion about 28 February,
    • Some people did overpay, they are being reimbursed, but contact organisers if there is difficulty with that.
  • The LOC produced a poster that can be downloaded (https://filesender.unina.it/filesender/?vid=48d79c11-8383-95c9-733f-0000585068b1) and printed.
  • Further follow ups on participation:
    • Liz was trying to encourage the head of LArSoft (Erica) to come for the neutrino field. Will try to get others (Tom, Heidi, Andrew).
      • David: Erica should be coming.
    • KEK/Belle II are coming, but could have more (Michel).
      • Thomas Kuhr has resent the email - should have ~3 key people.
    • BES/IHEP participation
      • Pete contacted and Xiaomei Xhang is coming and can give an IHEP report.

• Monday

  • Opening Plenary
    • Added IHEP talk from Xiaomei Xhang - need to follow up on content.
  • Technology Watch
    • Added 10’ to the slot to get the interesting network talk by Enzo Capone from GEANT (cuts 10’ into coffee).
    • Iterated with the organisers re. the vendor talks, especially RITALL looked too much like a sales pitch.
  • HEP Use Cases
    • DUNE reconstruction challenges : presentation now finalized.
    • Others still being discussed.

• Tuesday

  • Data Management/Lakes
    • For Data Management, both Plenary and Parallel are filled, speakers confirmed, agenda can be considered final.
  • Frameworks and Infrastructure
    • Session is now in Indico with the three confirmed talks. Just need to send few notes for the speakers to stress the points we would like to have in the session.
  • Training (Parallel)
    • Agenda ready in indico, due to tight schedule we provided slide preparation instructions to speakers. Might modify a thing here and there.
    • All speakers confirmed (including people from LIGO/VIRGO).
  • DPHEP (Parallel)
    • More details have been added to the draft agenda:
      • Certification: Motivation, Benefits and Status
      • CWP: Challenges facing LTDP in HEP
      • Data Access (& Discovery?) - A New Achilles' Heel in LTDP?
      • Collaboration Across HEP and with Other Disciplines: A possible EIROForum Sub-Group (following visit from ESRIN to CERN in February)
      • Archiving and Preservation "in the Cloud" - Outlook
      • Preparing Input to update of European Strategy for Particle Physics (and others elsewhere in the world)
  • Workload Management
    • For the workload management session we've defined the talks and will shortly start arranging speakers. Notebook.
  • Analysis Facilities and Use Cases
    • Agenda now finalised.

• Wednesday

  • Programming for Concurrency and Co-Processors
    • Session in good shape.
  • Visualisation (Parallel)
    • Many contributions have been confirmed by the speakers: ATLAS (Arturo), CMS-ROOT-GSI (Sergey), ATLAS (Ric, Ilija)
    • Also, contribution from University of Ljubljana (not in HEP) confirmed. The travel cost is funded by S2I2 in the context of a CMS-ATLAS-Ljubljana joint project to explore new techniques to be used in common tools, along the lines proposed in the CWP paper.
    • There will be a discussion on Vidyo with people who cannot attend in person (some from US).
    • Other contributions not confirmed yet.
  • Simulation
    • Added presentations from ALICE and LHCb, and one on the Geant4 R&D programme
    • Agenda is now very full - total of 13 talks. Some re-arrangement of order and timing of talks is still foreseen.
  • Security (Parallel)
    • Ian Collier volunteered to help run the session.
    • Timetable on Indico is essentially accurate
      • Speakers confirmed
      • Vincent connecting remotely
      • OpSec split to be discussed early next week
  • Performance and Cost Modeling (Parallel)
    • In good shape
    • Problem identified for the “Metrics & Measurement” presentation (neither Graeme due to overlap with SW development, nor Gareth or Michel can make it)... Last news: Gareth could give it remotely.
  • Technical Discussion on Frameworks (Parallel)
    • Looking nice, already in good shape
  • Software Development (Parallel)
    • Progressing, 5 topics will be discussed: packaging, profiling, static analyzers, project incubation, participation to open-source projects/conferences
    • Graeme notes that some software packages identified in EP R&D process could be good candidates for being HSF projects (tkLayout) - how do we do this?

• Thursday

  • Closing Plenary
    • Avoid summary of the summaries but rather some sort of “visionary talk(s)”

CWP

• General Matters and Roadmap

  • Michel gave a presentation about lessons from the CWP and future work at the 4th CERN Scientific Computing Forum, on Friday 23. Very well received, recognized as an important milestone. HSF encouraged to continue the work towards more commonalities, in particular for the frameworks. See email to HSF Startup list. WLCG preparing its strategic plan out of the CWP: early draft soon, final version in Spring, will have a strong focus on the data lake concept (also part of new European proposal, ESCULAP, presented during the SCF, in common with SKA).
  • WLCG will play a key role, especially for the computing evolution. HSF will play a strong role in the software developments to bring coherence.
  • CERN Courier article on HSF and the CWP is in proof with the printers now - looks great! Will appear in April edition.

• Publication strategy for Individual WG Papers

  • Simulation

    • Ready to go to arXiv now, with a finalised author list.

  • Machine Learning

    • Should chase this again.

  • Software Trigger and Event Reconstruction

    • https://arxiv.org/abs/1802.08640
    • https://arxiv.org/abs/1802.08638

  • Data Organisation, Management and Access

    • Michel to chase this again.

  • Data and Software Preservation

    • Should chase this again.

  • Data Analysis and Interpretation

    • It’s in good shape - Mark and Oli just reviewing the final version.

  • Visualization

    • Still editing, body is good, needs conclusions.

  • Event/Data Processing Frameworks

    • Should chase this again.

  • Careers, Staffing and Training

    • Want to incorporate thoughts from Naples and revise, given the wider community involved there.

  • Facilities and Distributed Computing

    • Waiting until after the WLCG strategy document is done.

  • Conditions Access

    • 2/3 done, lack of time to do the last 1/3...

  • Generators

    • No progress yet… lack of time

PyHEP Workshop

• Couple of sessions finalised.
• Organisation of presentations and speakers progressing. Will be easier to chase speakers once CHEP announces the accepted talks and posters, as the overlap of audience is going to be rather large.

Activity updates

Licensing

• Witek is following up about HepMC3 with Andy Buckley - hope to © to CERN and apply LGPL.
• Fastjet follow-up: after Naples

Packaging

• Next meeting in 7 March.

GSoC

• There is activity with prospective students now things are live.

---