HSF Weekly Meeting #160, 28 February, 2019
Present/Contributors: Graeme Stewart, André Sailer, Marko Petric,
Tommaso Boccali, Danilo Piparo, Pere Mato, Stefan Roiser, Andrea
Valassi, Eduardo Rodrigues, David Lange, Agnieszka Dziurda, Serhan Mete,
Andrea Rizzi, Caterina Doglioni, Ed Moyse, Jim Amundson, Liz
Sexton-Kennedy, Mark Hodgkinson, Daniel Elvira, Heather Gray, Markus
Klute, Gloria Corti
News, general matters
- LHCC WLCG
- Graeme presented slides from software projects and HSF.
- Question asked about Geant vectorisation R&D:
- Emphasised again the intent to deliver improvements into
- Discussion about accelerator integration and framework from
- Pointed to JLab workshop as a place to discuss this.
- Support from referees for common software work.
- AccelNet:FRESCO support letter
for project description and draft letter.
- To clarify, the HSF in the project application is described
- HSF: The HEP Software Foundation (HSF) encourages
cooperation and common development of software in HEP. The
HSF hosts a number of software-focused working groups and
projects where developers are collaborating. These include
groups on physics generator models, detector simulation
techniques, event reconstruction techniques and approaches
to software packaging.
- i.e. HSF as a forum to build networks and locate community
- André can you mention CLIC?
- Meeting approved the letter of support.
Google Summer of Code 2019
- We are accepted as an organisation again.
- There is a meeting for all mentors
tomorrow at 14h CERN
- Will cover how to interact with students and the timeline.
- Proposal for Advisory Committee:
- Please review and comment on the proposal in the next week, aim
to approve in next week’s meeting.
- If you would be willing to serve as on the AC please let HSF
HSF/WLCG/OSG Workshop News and Planning
- Indico registration is still open.
- There is a Google Sheet linked from the
page to help
coordinate car sharing from Dulles (you take your own
responsibility for this!).
- Reminder: David Lange, Michel Jouvin and Graeme Stewart organising
- General Sessions:
- Plenary sessions all fixed now.
- HSF Sessions:
- Will do a sound check with Amber and LOC next week to
verify Vidyo connection is acceptable.
- For planning other HSF Parallel Sessions, see WG activity
- We will contact Amber about JLab entry requirements (some worries
that people may need CVs due to changes in DOE rules).
Activity and Working Group Updates
- Draft agenda for JLAB discussed, here a draft (speakers being
- 1st session: Where are we now, where are we going ?
- 20’ talk on analysis needs and how they are served today
- 20’ talk on what’s missing today (Andrea)
- 20’ talk on evolution of HEP computing implications for
analysis (data distribution, access to resources, hardware
platforms, …) (TBD)
- 30’ discussion
- 2nd session: Technology
- 20’ talk on the future techs (Danilo)
- 20’ talk on Python and frameworks for declarative
- Check for overlaps with PyHEP session talk on Coffea!
- 20’ talk on ROOT Declarative framework: RDF (Stefan)
- 30’ discussion
- Timetable should go into Indico, even if not 100% settled.
- Agenda for the HSF meeting is finalised – both talks and
- Overall philosophy: tried to assign a balance of speakers from
experiments. Expect each speaker to summarise one topic
focusing both on their own work and also the perspective from
- Next topical meeting on 6 March:
- Need to add agenda entries for that meeting.
Reconstruction and Software Triggers
- Agenda for HSF meeting finalised (talks and speakers).
- Session 1: real-time analysis, talks from LHCb, ATLAS, CMS,
- Session 2: ML/accelerators for reconstruction, talks on LHCb
tracking, GPU for IceCube, ML in FPGA, JLab reconstruction
- Plan to have a chat/meeting with all speakers to prepare the
- We (WG coords) will look at the talks and prepare 2-3
questions in advance so that we can have a discussion at
- General direction of questions:
- Is there room for collaboration and where?
- Are upgrades & detectors real-time-analysis ready?
- Topical meetings: would like to have a shared calendar with other
groups (e.g., IRIS-HEP) to pick dates for our own without
- It’s already hard enough for us to find time where we’re all
- Should we have an HSF + others google calendar?
- This is possible with the current HSF calendar.
- Graeme will send instructions on how to edit and add
- JLab session well defined.
- JLab 1.5h session is finalised. Outside speaker from Anaconda.
- Will have regular open meetings every two-weeks, first is this
afternoon at 16h00 after this meeting (but on a different Vidyo
room). 28th Feb
- Discussion on Gaudi Plugin Service / Gaudi License with CERN KT.
- Re. code ©CERN, responsibility for re-licensing open source
code is given to the CERN group that develops the code. In
this case EP-LBC.
- Thus is should be fine to relicense from GPLv3 ->
- With that change the intention is to ©Gaudi to CERN, for the
benefit of the ATLAS and LHCb collaborations; and to apply an
- CERN have a webpage that recommends
- Still no response from FastJet authors.
- Held a meeting on 27
(Chris Burr attended).
- Spack update (HEP people now have merge rights for our
packages; we also have our own Slack channel).
- We got final proofs from CSBS.
- Eduardo and Michel noticed quite a few errors, which have been
pointed out to Springer and will be corrected.