Simulation Project Leaders Meeting
23-February-2010

==== Attendees

John Apostolakis (JA),  Alberto Ribon (AR), Witold Pokorski (WP),
Gabriele Cosmo (GC).

Excused: Pere Mato, Rob Veenhof.

---------------------------------------------------------------------
==== Minutes

- Extremely positive results for simulation in ATLAS and CMS with
  comparison with real data presented at the LHCC meeting on Wednesday,
  February 17th. See:
    http://indico.cern.ch/conferenceDisplay.py?confId=81149

---------------------------------------------------------------------
==== Simulation physics validation & framework - Alberto

- Meetings:
  o At the Physics Validation meeting on February 10th, M.Kossov
    has reported about the improvements implemented in CHIPS elastic
    cross-sections for hadron-nucleus interactions, involving the
    following hadron projectiles (h):
      p, pi+, K-, K+, pbar, Lambda, Sigma+
    and the following nuclei (A): 
      d, He, Be, C, Al, Cu, Sn, Pb.
  o The next meeting will be on March 31st and/or April 21st.
- Validation studies:
  o The activity has been focussing on microscopic checks on energy
    transitions between models, based on release 9.3. It was verified
    moreover that a problem in FTF_BIC of energy conservation has been
    fixed in this release.
  o Some issues have been identified in CHIPS in release 9.3 due to
    not optimal tuning applied in the attempt to speed-up the model.
    Fixes have been already provided and are now under testing, with
    the idea to include them in a future patch.

---------------------------------------------------------------------
==== Generator services - Witek

- Meetings:
  o There will be a MC4LHC meeting on March 29th to April 2nd, for
    LHC readiness of MC generators, where also GENSER will be
    represented.
- GENSER:
  o Introduced new version of Pythia-8.1.35 and a new generator,
    NLOJet++. Added also a new generator Tauola++, first Beta version
    of the C++ version of Tauola, aiming to use the new 'autotools'
    based build system of GENSER.
  o Request by ATLAS to port old versions of generators to SLC5
    (IsaJet, Photos). 
- Validation:
  o Putting more emphasis to validation and testing and organizing
    meetings on these topics. Established collaboration with the DESY
    group for adoption of the HepMC-Analysis tool.
  o Pushing to use Rivet for validation in GENSER. Proof of concept
    level tests for regression testing are being developed. Trying to
    also use the data sets available in Rivet for validation.
  o Started discussion for the Physics Center concerning the MC
    generators tuning and validation; GENSER will be involved in this
    from both the technical side and in the actual tuning and analysis
    from the physics point of view. A meeting on minimum-bias
    and underling-event is scheduled on March 1-2nd, and a section
    will be dedicated to tuning and validation.
- HepMC:
  o Planning meeting held 3 weeks ago; ideas for the features to be
    included in the new release 2.06 were presented by L.Garren and
    agreed. The new release is scheduled for beginning of summer.
  o It has been proposed to drop support for older releases 2.04 and
    2.05, as soon as the new release 2.06 will be out.
    Also proposed to add a quick migration guide in the release notes
    of 2.06 to help in the code migration from 2.03, the current
    version used by the experiments.

---------------------------------------------------------------------
==== Fluka - Alfredo

- No report.

---------------------------------------------------------------------
==== Geant4 - John

- Releases:
  o A new patch to 9.2 (9.2.p03) has been released on February 19th
    including a collection of fixes already part of 9.3; fixes are
    mainly related to memory use, initialization boost, and other
    minor ones.
- Developments:
  o Retuning of the parameters in the CHIPS model and cross-sections
    for kaons is being prepared; planned for inclusion in the next
    development release.
- Feedback from experiments:
  o ILC feedback on lateral shower shapes, under discussion.
  o LHCb expressed need for precision in cross-sections of kaons,
    identifying this as area where improvement is needed in Geant4.
    Also expressed requirement for precise description of
    multiplicities in hadronic reactions, particularly kaons and pions.
  o A report for the ATLAS simulation performance has been completed;
    benchmarks were made using Geant4 built-in capabilities and
    developments specific to ATLAS.
    Planning to make this report becoming an ATLAS internal note and
    an LCG note.
    Also continuing collaboration with ATLAS for studies related to
    the use of CPU caches with ATLAS benchmarks.
  o In touch with CMS for performance studies focussing on the use of
    mathematical functions.
- Meetings:
  o The SFT simulation group at CERN is now participating actively to
    the ongoing Geant4 Architecture Review, which is mainly focussing
    on performance improvements, maintainability, impact of new
    hardware and software technologies.
  o Reactivated the Geant4 performance task force, in collaboration
    with colleagues at FNAL and participation of people in the
    experiments and relevant user representatives. The first meeting
    taking place March 8th.

---------------------------------------------------------------------
==== Garfield - Rob

- No report.

---------------------------------------------------------------------
==== AOB

Next meeting scheduled for April 13th.