Present:
Allison Hall,
Andreas Salzburger,
Ben Morgan,
Caterina Doglioni,
David Lange,
Eduardo Rodrigues,
Graeme Stewart,
Josh McFayden,
Kilian Lieret,
Krzysztof Genser,
Kyle Knoepfel,
Liz Sexton-Kennedy,
Mason Proffitt,
Meirin Oan Evans,
Michel Jouvin,
Michel Villanueva,
Paul Laycock,
Serhan Mete,
Stefan Roiser,
Sudhir Malik,
Teng Jian Khoo
Apologies: Andrea Valassi, Dorothea vom Bruch
News, general matters, announcements
HL-LHC Review Phase 2
We have had two mini-workshops that presented plans and ideas for the HL-LHC era from both experiments and projects (a conversation):
These were well attended and extremely useful.
The inputs and discussions will inform the documents that are being written. First drafts to be ready 30 June, so that they can be discussed and get feedback over the summer.
Google Summer of Code
GSoC administrators have requested slots for all proposals where they received a good proposal (~27).
Proposal for DUNE Framework Requirements Review
Framework requirements review kick-off to be held next week (probably Mon. or Tues., time to be settled by the end of the week). Review mini-workshop will be during the week of May 24th or May 31st.
Working Group Updates
Data Analysis
Second instance of “Wider HEP/NP analysis community” meetings yesterday. Agenda with recording & live notes here.
- Excellent, interesting talks from FAIR/GSI, LZ, Xenon
- Highlights:
- Sharing of basic SW framework and dynamic computing resource allocation at GSI
- Effective Gitlab CI, common analysis SW in LZ
- 100% python framework at Xenon
- Put together our own “talking points”
- DUNE framework review – generalisable?
- How have (some, smaller?) experiments managed to converge on one analysis framework
- Keen to revisit discussion on recognition (always a popular topic!)
Need to define an agenda for the next instance, currently scheduled for 19 May but perhaps we should postpone it to a week later to avoid vCHEP…
Analysis metadata “constraints” rough draft written, to share with community for comments soon:
- Already solicited comments from Paul (thanks!)
- Feel free to add yours
Detector Simulation
- Next meeting 10th May on simulation for solid state microphysics:
- Provisional next meeting 7th June on pyg4ometry (see also https://arxiv.org/abs/2010.01109) package for geometry construction
- Likely of interest to Reconstruction/Triggers and Python WGs? (literally set it up a couple of days ago, so not advertised yet!).
- May also have an update on GDML parsers/support in Geant4/ROOT/VecGeom
- Preliminary topics for fall: MARS, varying Geant4 parameters
Reconstruction and Software Triggers
- First topical meeting on timing reconstruction (LHC + Future Expt) has a lot of interest. Trying now to find a date.
- Open data detector + timing files are almost ready for release.
PyHEP
- Topical “meeting of the month” on the Dask package held yesterday Wednesday, see https://indico.cern.ch/event/1027094/ for the tutorial materials (recording soon to be uploaded to YouTube).
- PyHEP 2021 workshop:
- Details on the Indico agenda.
- Reminder - registration and abstracts submission are both open.
- Seeing again a very high level of interest :-).
We had a WG meeting last Wednesday (Wednesday 28 April 2021, 5pm CERN time):
- Agenda: https://indico.cern.ch/event/1026182/
- Topic: Experience with Spack at Fermilab
- Speaker: Stephen White (Fermilab)
- Pretty good turn out with almost 80 people from various experiments/labs. Would be good to do a “post-mortem”, converge on a few action items based on the discussions and follow-up from there.
Software Training
- Modification in the web design of the Training Curriculum.
- A more intuitive interface. Modules are shown in boxes instead of lists.
- Instead of “beginner, intermediate, advanced”, the modules are now organized in collections such as “Basic”, “Software dev”, “Machine learning”.
- Easily maintained with YAML.
- A Hackathon follow-up was held on Apr 30 to discuss the status of the training modules towards the main Hackathon in June.
- For the next time, meetings dedicated to each module can work better than a general call for a follow-up.
- We need to define a list of the high-priority modules to develop/improve. Based on the list, we will organize the meetings.
- Some of the modules require input from experts in other WGs. We will ping the contact person when required.
- As an example, for Singularity we are working in a chapter that aims to deploy Pythia8 in a container.
- Discussing when the next round of HSF training events must be held.
- Candidates:
- Summer, when people are usually with more time available.
- Sep-Oct, when new PhD students take place.
- Improve contact between training WG and other WGs
- Nominate a training contact for each WG - WGs please follow-up on this
Event Generators
- Started series of meetings to gather feedback for LHCC Review 2:
- Had a very productive meeting with Sherpa authors on 22nd April
- They showed a number of CPU improvements recently implemented and/or in the pipeline
- We discussed some very interesting “bigger picture” ideas such as creating a “global” modular plug-and-play generator interface
- We have an HSF Generators WG meeting this afternoon:
- LHCC Review 2 input from EvtGen (Photos also invited to contribute).
- Update on MadGraph5_aMC@NLO migration to GPUs
- The next meeting will probably be on May 20th (Pythia)
- Then one on May 27th and others from early June (Powheg, Madgraph, Herwig to be scheduled precisely).
- Has been a bit hard in places to pin people down!
Frameworks
Monthly informal “coffee” delayed one week until May 12.
AOB
CAF and Roundtable
Webpage
- Cannot properly deploy to github pages from a fork of the repository (links won’t properly resolve)?
- This seems to be related to the usage of absolute links (e.g.
/css/hsf.css) without using `` (which only works if everything is directly relative to the domain, as with the official deployment to hepsoftwarefoundation.org, as / resolves to that)
- Some PR with design tweaks:
- Slightly increase overall font size:
- Add some icons to Nav bar
- Boxes with “big” links to make the important things stick out
Next Meeting(s)
- Next meeting will be 27 May (after Ascension and vCHEP)