Simulation Project Leaders Meeting 27-September-2005 ==== Attendees John Apostolakis (JA), Paolo Bartalini (PB), Alfredo Ferrari (AF), Witek Pokorski (WP), Alberto Ribon (AR), Rob Veenhof (RV), Pere Mato (PM), Gabriele Cosmo (GC). --------------------------------------------------------------------- ==== Minutes The final Application Area planning document for LCG Phase II has been finalised and is available from the AA web site. It also includes the Simulation Project planning for 2005/6. --------------------------------------------------------------------- ==== Simulation physics validation - Alberto - Tile calorimeter test-beam: + A meeting with the people involved in the original first study for the Tile calorimeter has taken place, the new production should start now using the new setup prepared by Manuel Gallas. Once this work will be completed (in October), it will be possible also to verify the performance of the more recent versions of Geant4 concerning the known problem on the hadronic shower shapes. - 3rd simple benchmark: + The setup has been implemented and is now running both for Geant4 and Fluka. For Fluka, the new version of the software is used. + First results have been obtained and now the physics needs to be understood with the simulation experts. It is also required a better understanding of the data used to identify possible contaminations affecting the comparison with the MC results. - There has been recently contacts with Daniel Elvira (CMS) concerning CMS studies on low-energy hadrons data. A presentation on this work is planned for the next validation meeting in October. - Still work going on in ATLAS on the combined calorimeter. It is expected a presentation at the next validation meeting in October. - A new fellow, Alex Howard, will join the project starting from October 3rd. He will concentrate on radiation studies continuing the work initiated by G.Daquino. --------------------------------------------------------------------- ==== Generator services - Paolo - Meetings: + LCG generator meeting held in July where it was announced the new release of GENSER and the alpha version release of Pythia-8. + An open meeting on Herwig++ was held in August. + At the September 29th meeting it will be presented a status report on GENSER and a demonstration of MCDB. - Manpower: + Current shifters are Alexander Toropin (GENSER), Sergey Belov (MCDB/HEPML) and Anton Gusev (MCDB/HEPML). + A new INFN fellow (Marianne Bargiotti) just joined the project and is currently working on WP4 (Monte Carlo Validation) - Milestones: + The creation of the MCDB user document with definition of procedures to gain access through certificates is sligthly delayed but progressing. + The release of the alpha version for Pythia 8 has been achieved and Beta testers are working on it. + Concerning the definition of EvtGen development plans & policy (agreement on responsibilities for EvtGen development in GENSER), the situation is evolving but is not yet established. + Concerning the procedure for light GENSER releases and packaging, the next GENSER release scheduled for October 20th will use the new procedure. Documentation on the new packaging scheme must be provided and it should be verified with the experiments that the requirements are met. It is expected that Andreas Pfeiffer will decrease his contribution to the project in future. This may open a problem that needs to be sorted out. + First introduction of NRQCD Prompt Quarkonia Production models in Pythia 6.3 has beem achieved. - GENSER: + Release 1.1.0 has been made available end of July; tests have been moved in separate sub-package. The documentation has been improved. New tools are in place: python scripts for package installation/compilation, check logging system and post-installation checks. + There's on going activity for updates of packages and testing/porting on new platforms. - MCDB, HEPML: + Improved usability and user interfaces. Test with large files was successful. Insertion MCDB to support public generator level files. - Validation: + Test packages are growing (Alexander). NRQCD model recently introduced in Pythia 6.3 is now under testing (Marianne). - Simulation Framework: + The prototype was reviewed and works but it had not really followed the design. In parallel with the reorganisation of the software in CMS, the project was restarted to follow a more coherent vision; there will be a task force in October working on it. It is however not clear at what level the project can be driven in LCG, given the limited contact with the people involved and the different objectives more CMS-oriented. - Concerns are expressed about the future; perspectives are not clear: it is intention of LCG Russia to reorganise priorities and strategies, and there's serious lack of MC experts in the project. It is required to clarify which resources will be available in the project from now and 2007 for achieving the LCG-phase II milestones which have been established. --------------------------------------------------------------------- ==== Fluka - Alfredo - The new release of Fluka: + Has been accomplished end of July. It includes technical improvements and also updates in the physics. In particular, for what concerns mainly online radioactive decays, the whole process for generating the radiation profiles in decays has been integrated on-line. This has several technical advantages: savings in time, safer since less exposed to user interventions and allowing to make a proper stastical analysis of the results. This part has been already benchmarked by the radio-protection group; one of the advantages of the new approach is that the mathematics is now able to manage whichever nested level generation of radioactive nuclei. + The new Fluka can be downloaded from the web; binaries can be downloaded freely as usual, while sources can be distributed only for CERN and INFN people and written acknowledgement of the license is necessary. + The documentation has been completed end of July, waiting for approval as yellow report which is expected by October. + There have been some fixes included in the last packaging, mainly related to the handling of radio-active nuclei. An incremental bug-fix release is expected in one month including fixes for relativistic heavy-ions. + There will be a collaboration meeting in Milano mid-October where also participation to upcoming conferences and publications will be discussed. - There is increasing involvement in the hadron therapy field, this implies refiniments to the models for ion interactions (at few 10ths MeV per nucleus). The collaboration is mainly happening with the German Center for hadron therapy; they're mainly interested in PET monitoring of the therapy (including Beta+ emitters), since it is now also possible to handle CT-scan geometries. - Installation of the libraries at CERN will be done following the LCG policies. - It is still needed to verify the porting on 64-bits Itanium architecture, for which access to a system, eventually at CERN is required. The new Fluka runs on Opteron in 32-bits mode, no big advantages were observed in 64-bits. --------------------------------------------------------------------- ==== Simulation framework - Witek - Geant4 geometry object persistency: + The feasibility study for performing object persistency of Geant4 geometry has been completed. All the elements necessary to achieve the full functionality have been identified and a solution for applying the necessary changes in Geant4 has been agreed; the modifications will be introduced in due time in the Geant4 code. - GDML: + The Python implementation of the GDML writer has been completed for Root. For Geant4 it was decided not to proceed with the full implementation, but eventually define a Python layer making use of the existing C++ implementation. New release GDML-2.3.0 announced on August 3rd. + The foreseen extensions to GDML for supporting divisions/reflections and files modularization have been postponed due to lack of manpower; no technical student has been yet assigned to the project. - MC-truth + A special meeting on MC-truth handling is happening on September 28th with experiment representatives, the main purpose of the meeting is to review the various approaches adopted in the experiments and eventually identify ideas and solutions of common interest. - Flugg: + The work for the Tile test-beam validation has started. The latest version of Fluka is used; a skeleton is ready and working, now starting doing physics with it. --------------------------------------------------------------------- ==== Geant4 - John - Milestones: + For the contributions to the developement release of September, one of the key items was the study on the stability of EM quantities from sampling calorimeters against changes in cut; the problem which was first observed from user studies for medical and space applications, and then confirmed in HEP studies. The underlying case has been identified and the study has been completed. Refinements and improvements in Multiple Scattering have been undertaken and a development version of the relevant revisions is being included in the September development release. + Concerning improvement to the regression suite for release validation and testing infrastructure, due in October, this is proving challenging and is likely delayed. Effort has been concentrated in implementing the recent change in coordination of the primary testing activity, which is now undertaken by Ian Mclaren. - Fixes and developments: + A problem was reported recently by experiments using fast-simulation, where electrons were deviated before it could apply itself. After consultation with the Geant4 team, a user in ATLAS undertook an extensive investigation and found the reason of the problem and proposed a valid fix for it. The fix is currently under testing and will be eventually included in the September development release. + In the geometry, a fix to handle complex sphere shells has been introduced and will be included in the September development release. + There is ongoing support to enable experiments to simulate exotic particles simulation (l-hadrons, monopoles, ...). + A design iteration is proceeding to allow coupled navigation in two or or more geometries in parallel. This will support the use cases of LHCb and others, in which tracking in field, biasing and tallying (and/or fast-simulation, ...) each have their own geometry, and charged tracks proceed in lock-step through these. + Physics lists: technical issues and physics options are under investigation. + Data: a cataloguing and first assessment of the external data libraries used is ongoing, included a research for the conditions of use of the latest and historical versions of some data libraries. - The new Collaboration Agreement has been sent to the Collaboration Board (CB) members for the final endorsement. - There will be the annual Geant4 Collaboration Workshop in Bordeaux beginning of November, which will also include the public Users' Workshop. Progresses on the issue of the license are expected as well. --------------------------------------------------------------------- ==== Garfield - Rob - Milestones: + Participation to the TRD-2005 conference in Ostuni with the opening talk. It was the occasion to assess the precision which with photo-absorption cross-sections and fluorescence yields in argon and xenon are known. This clarifies the miss match between Geant4 simulations of the CAST TPC efficiency vs photon energy. Except in the case of xenon, where more studies are needed, Heed, and in particular the latest release has been shown to be highly reliable. + The interface with the new version of Magboltz has been completed. The new release is available, includes anisotropic cross-sections for all noble gases as well as CS2. The documentation of the cross-sections has been brought up to date. - Note: S.Biagi, the author of Magboltz has retired. The problem for future support of Magboltz needs to be sorted out. --------------------------------------------------------------------- ==== AOB Next meeting end of October.