Notes from Geant4 Steering Board #1, Catania, 17 Oct 2009

Version: 1.4, 10 Dec 2009, 12:25 JST

Editors: J. Perl, J. Apostolakis


- John Apostolakis (chair)

- Takashi Sasaki

- Sunanda Banerjee

- Makoto Asai

- Giacomo Cuttonne

- Pablo Cirrone

- Gunter Folger

- Hisaya Kurashige

- Dennis Wright

- Vladimir Ivantchenko

- Peter Gumplinger

- Joseph Perl (notetaker)


- Katsuya Amako, Gabriele Cosmo, Sebastien Incerti

1. Preparation of Geant4 2010 Architecture Review

The proposed action for 2010 is an internal review of the architecture

of Geant4. The outcome of this review should be a prioritized set of areas

for redesign, with effort estimates.

Redesign work would be undertaken as a result of the review.

External experts to be invited as needed.

This will be a review of priority areas:

- key interfaces, in particular those between categories (including Process)

- external interfaces, including all dependencies

- infrastructure classes including exception

- assessing the use relationships between categories (reduce crosstalk)

- impact of addition of ‘major’ functionality since 1997

Proposed timeline:

Review to report in time to use the 2010 Collaboration Meeting,

to plan the architecture revisions.

Potential targets: CPU performance, maintainability.

To consider technology shifts such as multithreading, multi-core, grid.

2.) Release 9.3

To complement the list of developments (presented in workshop)

a draft list of areas at risk was created.

Examples which cannot are not working in time should be not be included of

the release.

No cross-working group issues regarding developments.

Related Topics

a) Physics revisions

Identified risk of changed physics behavior within a well-known physics-list

name (e.g. QGSP_BERT)

Adding more options is preferred (until improved results are well established),

so long as the changes are well documented.

Concern that validation results are communicated to enable collaboration

decision on choice for production physics lists.

b) Identified need for improvement of a number of advanced examples.

Planning to request authors to confirm their continued involvement in

maintaining the examples, and to help them make improvements.

Agreed action(Pablo): Identify examples which required code review and

invite category coordinators to arrange a review of the parts corresponding

*to their area of responsibility.

  • Consequent action (Category coordinators): Identify persons with expertise

    to review the code in the list of examples corresponding to your category.

3) Progress with the Review 2009 Responses

Report of the 2009 Review, including recommendations has been made public.

It is linked to the agenda of the Review.

Clear proposals from responsibles not yet available for each recommendation

identified as first priority. A decision from steering board to adopt or not each

of these proposals will follow.

Confirmed that existing planning of hadronics is considered in its present form

- it does not need to be copied (duplicated) in additional documents.


(a) Proposal for parts of GAMOS to be included in ‘environments’ directory.

The proposed target, the environment directory has a different mandate,

to enable simple interactivity with existing Geant4 functionality (not to extend it).

Given that the proposed package included a large number of additional

developments, the environment directory was not judged appropriate.

The revised proposal to include some of the functionality directly as part of

the toolkit was not judged appropriate to discuss at the SB, as consideration

first by the relevant WGs is required.

(b) Giacomo Cuttone reported that the paper on PIXE published in

NIM B (H. Abdelwahed et al.) in January 2009 was sent for publication

at INFN’s responsibility, without consultation of Geant4 Steering Board.

(c) Peter: cross issue between physics lists and generic processes.

There is an order in which processes have to be in process manager.

Nothing in Geant4 checks that this is done correctly.

This can be seen as a point for potential improvement.

Notes from Geant4 Steering Board #2, Catania, 22 Oct 2009

Version: 1.1, 22 Oct 2009, 17:00 CET

Editor: G.Cosmo

1) Actions

#42.3: GF will introduce test in next week

#51.3: Page exists; can be published. To be closed.

#51.4: Still open.

#51.5: Still open. CPU benchmarks.

#52.1: Still open.

#53.1: Some lists are still to be added. Still open.

#53.2: Still open.

#54.1: Document is available on SB meeting of July 2006. Still open.

- Should draft a statement on the incredible effort made by EM and HD

developers to improve the status of physics in past months.

2) 2009 Geant4 Review actions

- Updated Google document.

- New action #55.1: Request multi-thread team report and proposal in 2010/Q1.

- New action #55.2: Cat.Coordinators to identify responsible for documentation

or physics models whose original authors have departed Geant4.

- Proposal and plan for the replacement of the build system with the adoption

of cmake, approved.

3) SVN

- Not enough information is available at this time concerning the support for CVS

and having an update in the next 2 months for more details on the evaluation of

SVN, for approving the migration.

4) Geant4 census(es)

- February 14th proposed date for completing the census.

- M.Asai volunteer to help in the preparation process.

5) Elections of SB representatives

- Proposals for coordinators and deputies are approved by the SB.

The following proposals for coordinators and deputies, resulting from working

group elections are approved by the SB:

o Paul Gueye as the additional representative from Low-E

o Dennis, Gunter as Hadronics WG coordinator and deputy, and

Sunanda Banerjee and Giacomo Cuttone as the additional representatives.

- Target date for the new election is fixed for March 30th.

6) Election of spokesperson

- Start date for the process, April 15th.

7) 2010 Architecture Review

- Architecture team: M.Asai, J.Apostolakis, G.Cosmo, H.Kurasige; replacement

for physics representative requested. Asked to physics WGs coordinators to

propose a joint candidate.

- Proposed plan approved.

- Possible date for a mini-workshop in June 2010 or fallback May 2010;

may coincide with an early date for a Beta release, as for 2009.

- Working group should report to every SB meeting, final proposal to be

distributed 4 weeks before mini-workshop.

8) AOB

- Schedule for release 9.3 approved.

- The next Space Users Workshop will take place in Seattle (USA) on

August 18-20, 2010.

- Discussed response of the OB to the letter sent by M.G.Pia.

Meeting Date

Thursday, October 22, 2009 - 12:00