Version 0.3, 4-May-2004 CEST, JA
CERN: J.Apostolakis (chair), G.Cosmo, A. Dell'Acqua, G.Folger, V.Ivantchenko,
M.G.Pia, F. Ranjard, S.Sadilov.
SLAC/California: M.Asai, J.Perl, D.Wright and R.Mount (in part, only for CA item).
KEK/Japan: T. Sasaki, H.Yoshida
ESA: G. Santin (deputising P.Nieminen)
R Mount recounted the reactions he had received from CB members and in particular representives of expected signing parties for the future Collaboration Agreement. He outlined the areas where concerns existed, and confirmed that it was his understanding that all partices agreed broadly with the spirit of the proposal (CA draft 5.12) of May 2003, and had concerns regarding a small set of issues: primarily on the composition of the future Steering Board, the role of the future Oversight Board. He outlined a revised draft that he had created, and proposed to circulate it to the TSB.
It was agreed that Richard will circulate the current draft documents (CA 7.01, CR 2.0) to the TSB in a matter of days, after potentially making a few small changes.
Action 27.A1 (J. Apostolakis): Organise extraordinary TSB meeting in 3-4 weeks on the subject of the revised Collaboration Agreement proposal.
The process to agree a future Geant4 license was discussed. It was agreed that it is very important to put in place a Geant4 license that will be mandatory for all future contributions, and (to the largest degree possible) will cover also contributions already made .
Action 27.A2 (J. Apostolakis) Undertake process to identify who are the license holders for Geant4, to the best degree achievable.
New requirements and requests from the 3rd proto-meeting of the Technical Forum were discussed. The planning to address items was reviewed and compared to the workplan of 2004.
Action 27.A3 (G. Cosmo / J. Perl) Address TF request 0315 to "Fix problems with the boolean processor": review current situation and recommend solution.
SLAC agreed to bring out Geant4 CD on a best effort basis. Close action 22.4
HARP reported that Monte Carlo production is ongoing, with 2.6M event at 8.9 GeV/c and 12 GeV/c produced for Be and Al targets. Four production centers are in use. Verified that event history is the same for the same random seed.
Use of Linux RedHat 7.3 and gcc 2.95.2 is planned as sole version in 2004, and support for this 'platform' for releases and patches before December 2004 requested. (See 'Platform Decisions' sub-section below for TSB decision.)
CMS reported 18 million events simulated. Planned to go to 30
M foreseen in the
Failure rate observed is 1 crash per 5000 events. Also 3 Heavy Ion events (50k-80k primaries) ran in 1 test job. Work is ongoing for the next big production to start in summer '04 (around 100 M events).
ATLAS reported they have run several preproduction tests (in preparation for
Challenge scheduled for April) in order to demonstrate easy running with G4 in a DC, to show that a G4-based simulation is affordable (in terms of CPU/memory/disk) and to verify that the program is robust enough.
After initial setting up problems due to a couple of bugs in hadronic physics that Hans-Peter & team promptly resolved and a problem with the ATLAS geometry, the whole production went through smoothly and succesfully (100% of the jobs completed). As of this meeting cannot quote a failure rate. CPU needed was within a factor 1.5 wrt G3, as was the event size (not depending on G4).
"It goes without saying that we are extremely satisfied with G4 and how it performs."
For the immediate future, ATLAS priorities are: i) stable performance and
robustness, especially for what concerns physics; ii) performance improvements,
in particular for EM physics.
"ATLAS would like to thank the Geant4 collaboration for the support and the commitment that we could appreciate in the last few (for us hectic) months. Thanks folks!"
LHCb reported that they build Geant4 as a CMT project to keep track to modifications applied to the code between 2 releases. Also build Doxygen documentation, to link into Gauss docs.
Concerning reproducibility of events with Gauss simulation: it was lost using Geant4 6.0.ref01, but recovered using the subsequent patch Geant4 6.0.p01. Sought assistance to choose version to use for the Data Challenge 04 (April-May); will try to use the latest release after verifying robustness, which is essential.
TSB agreed that releases up to the scheduled June release and patches
to this will support gcc 2.95.2. The next full release, 7.0 in December 2004
will NOT support gcc 2.95.2. (This is due to the age of the compiler, leading
to missing features and even the need for back-porting of code with newer features.)
Support for Windows Visual C++ 6.0 will be phased out in 2004. Users shall be notified in advance of tentative plans and consulted regarding the proposed end-date according to standard practice.
Action 27.A4 (G. Folger) Circulate proposed end-date for Windows Visual C++ support, and seek user feedback.
Priority on identifying problems and improving performance in 'hotspots' and modules with significant CPU use.
Agreed also importance of User Workshops and Tutorials.
Action 27.A5 (G. Cosmo) Will improve promotion of future Events for users in Geant4 home page and maintain separate page listing further details of upcoming and recent Tutorials and User Workshops.
Assess status of Toolkit Developers Guide, and effort required to maintain it, in comparison with its use in the community.
Action 27.A6 (D. Wright) Provide concrete proposal on potential reorganisation of Documentation to CC and then TSB.
MG Pia reported on the status of pre-planning meetings for this proposed project. It was agreed that a more formal project proposal, with manpower estimates was necessary.
Action 27.A7 (M.G. Pia) Provide a first estimate of the effort required for a Geant4 Physics Book Project to the TSB.
Action 27.A8 (Category coordinators) Provide information on deliverables for the March release to the release manager.
Action 27.A9 (G. Cosmo) To create a web page with features of the 2004 scheduled releases and make it available to users.
Action 27.A10 ( JA / MGP ) Propose Program Committe for workshop.
Action 27.A11 (Program Committe chair) Identify focal points for workshop.