Title: Activity report 2001 Plans 2002 http://www.ge.infn.it/geant4 http:/www.ge.infn.it/geant4/LowE/
1Activity report 2001Plans 2002http//www.ge.inf
n.it/geant4http/www.ge.infn.it/geant4/LowE/
- Maria Grazia Pia (INFN Genova)
- for the Geant4-INFN Group
Commissione Calcolo Roma, 13 December 2001
2 The Toolkit
- OO Toolkit for the simulation of next generation
HEP detectors - ...of the current generation too
- ...not only of HEP detectors
- already used also in nuclear physics,
astrophysics, space science, medical physics,
radiation background studies etc. - It is also an experiment of distributed software
production and management, as a large
international collaboration with the
participation of various experiments, labs and
institutes - It is also an experiment of application of
rigorous software engineering methodologies and
Object Oriented technologies to the HEP
environment
3Geant4 Collaboration
- MoU based
- gt100 physicists and computer scientists
- Distribution, Development and User Support of
Geant4
- Atlas, BaBar, CMS, HARP, LHCB
- CERN, JNL, KEK, SLAC, TRIUMF
- ESA, INFN, IN2P3, PPARC
- Frankfurt, Barcelona, Karolinska, Lebedev
- COMMON (Serpukov, Novosibirsk, Pittsburg,
Northearstern, Helsinki, TERA etc.)
- Collaboration Board
- manages resources and responsibilities
- Technical Steering Board
- manages scientific and technical matters
- Working Groups
- do maintenance, development, QA, etc.
Members of National Institutes, Laboratories and
Experiments participating in Geant4 Collaboration
acquire the right to the Production Service and
User Support For others free code and user
support on best effort basis
Budker Inst. of Phys. IHEP Protvino Univ.
Cordoba Northeastern Univ. MEPHI Moscow
Pittsburg University
4Status and activities
- 1st production release Dec 1998
- Since then
- 2 public releases/year
- monthly internal release
- Collaboration-wide objectives WG projects every
year - User support worldwide
- Well established in most major HEP experiments
(and also in small scale ones), space, medical
etc. - see CHEP 2001 presentations
Main areas of activity
- Kernel packages
- mostly in maintenance mode
- Geometry
- improvements, optimisation, XML etc.
- Physics
- very active development
- OOAD
- Interactivity visualisation, (G)UI, analysis
- active development
- Parallelisation, GRID
- hot topic
- Testing and validation
- internal activity collaboration with users
- Software process improvement
- significant investment and activities
5Il gruppo INFN in Geant4
- Cosenza, Genova, Milano, Torino, Trieste, Udine
- 20 fisici/tecnologi (molti giovani!)
- Principalmente in Low Energy Electromagnetic
Physics - Architettura, OOAD, Software Process, User
Support, sviluppo - Supporto delle esigenze degli esperimenti INFN
nella Collaborazione - Riferimento nella comunità INFN
- Esperienza di Geant4
- Esperienza di moderne tecnologie di software
(metodi OO) - Esperienza di metologie avanzate di software
engineering (software process) - Riferimento a livello internazionale
- Fisica di basse energie
- Esperimenti di astroparticelle / applicazioni
spaziali - Applicazioni bio-mediche
- Trasferimento di tecnologie ad ambiente medico e
spaziale - Stretta collaborazione con CERN e ESA (primo
contratto ESA allINFN!) - Eccellente palestra per giovani (e non solo
giovani) di approccio alle nuove tecnologie
6Areas of activity of the INFN group
- LowE Electromagnetic Physics
- Coordination (gt50 members)
- Physics, OOAD, testing, user support etc.
- Geometry and tracking
- Analysis
- Collaboration of frameworks
- Analysis tools for physics and testing
- Testing and validation
- Test tools
- Tests, comparisons with experimental data
- Advanced examples
- testing and user support
- OOAD
- electromagnetic physics
- analysis
- testing
- Software process and SPI
- Training
- Technology transfer
- Radiotherapy
- Radiodiagnostic
- Outreach
- Close collaboration with CERN
7How we operate
Goal-directed project management (GDPM) Rigorous
approach to software engineering High dynamism
and creativity
Characterized by
Physics Applications
Wide spectrum of development
- Regular Working Group meetings
- Emphasis on training of all group members
- Promotion of cross-WG activities
- Close relationship with user communities
- Active strategy of talks and publications
- for promotion of Geant4 and our own products
- as an incentive and reward for group members
Team
Collaboration
Outreach
8LowE e.m. physics developments
- Before the creation of the WG
- First implementation of e/photon models based on
LLNL libraries (250 eV - 100 GeV)) - First implementation of models for hadrons and
ions - Since the creation of the WG
- New model for positive charged hadrons (Ziegler
1985) - Improvements on straggling (hadrons and ions)
- Dynamic dependence on effective charge in the
range calculation - Improvements in the validity range of d ray
production - New models for ion energy loss fluctuations
- New model for negative charged hadrons (Barkas
effect) - New model for polarised Compton
- Various other physics extensions/improvements
- Future
- Polarised processes conversion, Rayleigh,
(photoelectric?) - New models for Compton, Photoelectric
- Positrons
- Penelope-like models
- PIXE
Compton Photoelelctric Rayleigh GammaConversion eI
onisation Bremsstrahlung hIonisation PolarisedComp
ton Atomic relaxation
To learn more http//www.ge.infn.it/geant4/lowE
9Results
This is a small sample of comparisons with
experimental data
Based on EPDL97, EEDL and EADL evaluated data
libraries
e,? down to 250 eV (EGS4, ITS to 1 keV, Geant3
to 10 keV)
Barkas effect models for antiprotons
Low energy hadrons and ions models based on
Ziegler and ICRU data and parameterisations
10Polarisation (TS/Cordoba))
Cross section
- Integrating over ?
- Sample ?
- ? - Energy Relation ? Energy
- Sample of ? from P(?) a (b c cos2 ?)
distribution
Sample Methods
Low Energy Polarised Compton
250 eV -100 GeV
Scattered Photon Polarization
More details talk on Geant4 Low
Energy Electromagnetic Physics
- ? Polar angle
- ? Azimuthal angle
- ? Polarization vector
Other polarised processes under development
11OOAD
12(No Transcript)
13(No Transcript)
14(No Transcript)
15Design of a generic user application
reverse engineering most of Geant4 categories
16Testing
Posted on the WG web site
- Suite of unit tests (extended 2001)
- 3 system tests(extended 2001)
- Suite of physics tests (in progress
with publications) - Regression testing (manual)
- Physics validation
- Testing requirements
- Testing procedures
- Testing integrated with development
- Validation
- Comparison with experimental data
- At various levels details of physics models and
global features
17Design of generic unit test for physics processes
exploratory, evolutionary prototype
18Torino testing and physics validation
Comparison of Geant4 e.m. physics with
experimental data
2002 plans for similar tests concerning proton
interactions too
19PROBLEM Verify Geant4 tracking accuracy of a
particle in a magnetic field.
Test configurations that optimise the elaboration
time and relative tracking error
Intersection points
Test geometry
Ending pointmomentum and field inversion
Starting point
Milano
Testing and optimisation
A set of configuration parameters has been found
getting errors of order 0.01?m and optimizing the
computing time.
20Advanced examples
- Three advanced examples developed by the LowE WG
Dec 2000 improvements 2001 - X-ray telescope (ESA)
- g-ray telescope (TS/UD)
- Brachytherapy (GE-IST)
- Full scale applications showing physics setting
guidelines and advanced interactive facilities in
real-life set-ups - New in 2001
- X-ray fluorescence and PIXE (GE)
- underground physics (UK)
- Plans 2002
- Underground background
- Hadronic physics in various energy ranges
- IORT, treatment head
- (subject to TSB approval and prioritisation)
ESA
TS-UD
digis, trigger
GE-IST
21TS-UD GammaRayTel
- Full geometry description
- CsI Calorimeter
- Silicon strips Tracker
- Plastic Scintillator
- Integrated analysis tools
- Trigger implementation with digits
- Visualization
- OpenGl
- Postscript
- VRML
22(No Transcript)
23- Geant4 allows us to define
- the geometrical structure of set-up
- describe the propagation of beam taking into
account the interactions through exit windows,
scattering thickness, monitor chambers,
collimator. All these materials, which are used
to shape and monitor the electron beam, will
modify position, size and energy of the produced
beam.
Air
Percent Depth Dose
Air Applicator (Lenght 80 cm,Radius 4 cm)
Simulation Set-up
Air Applicator Target (0.3mm Al)
24Use cases, OOAD, code, traceability in CVS
repositories
already done, to be refined
still to be done
Major role of analysis internal to Geant4
Analysis in Geant4
25Interface to analysis tools
Through abstract interfaces
? No dependence ? Minimize coupling of components
Example AIDA Analysis Tools
Anaphe
26Previous Geant4 analysis design
Reverse engineered No design existing
(confusing for users)
G4ANALYSIS_BUILDG4ANALYSIS_BUILD_LABG4ANALYSIS_L
AB_VISUALIZATIONG4ANALYSIS_BUILD_LIZARDG4ANALYSI
S_BUILD_JASG4ANALYSIS_NON_AIDAAIDA_DONT_USE_STD
G4ANALYSIS_USEG4ANALYSIS_USE_LIZARDG4ANALYSIS_US
E_LABG4ANALYSIS_USE_JAS
A lot of environment variables required
27No environment variables needed
or HistogramManager
New, simpler design Implemented in the advanced
examples for the release
More general and elegant, but still simple,
design under discussion
28Software Process
A rigorous approach to software engineering
- Huge effort invested into SPI
- started from level 1 (CMM) in LowE e.m.,
analysis, examples - chaotic, left to heroic improvisation
- Current status
- public URD
- full traceability through UR/OOD/implementation/te
st in CVS repository - testing suite and testing process
- public documentation of procedures
- A sample of the improvements
- establishment (1999) and management of User
Requirements Document - establishment and documentation of procedures
- design before coding
- designs and testing under configuration
management - separation of examples from tests
- identification of tasks and responsibilities
- traceability process
- ...too long to list all of them
Plan for 2002 RUP-based process (customization
of RUP to HEP)
29Technology transfer
- Istituto Nazionale per la Ricerca sul Cancro (IST
Genova) - AIRCC (Ospedale Mauriziano Candiolo)
- Ospedale di Savona
- Azienda Ospedaliera di Cosenza
- Istituto Nazionale Portoghese per la Ricerca sul
Cancro, Coimbra - Karolinska Institutet, Stoccolma
- Vari ospedali francesi coinvolti nel progetto di
PET associato alla Crystal Clear Collaboration - Centre Hospitalier Universitaire de Québec
- Massachussets General Hospital
- General Electric
Brochure (in collaboration with CERN) Article on
CERN Courier Seminar (Science and Society series)
Outreach
30Collaboration with user groups
Pixel ionisation chamber Relative dose in water
? Independent validation of physics models and
implementation
INFN Torino
Some examples
Photon attenuation coefficient in water
LIP Portuguese Oncological Institute
Italian National Institute for Cancer Research
31Presentations 2001
- Geant4 Low Energy Electromagnetic Physics, M. G.
Pia, Round Table on Monte Carlo Methods for Space
Applications, ESA/ESTEC - Geant4 Low Energy Polarised Processes, G. Depaola
and F. Longo, Round Table on Monte Carlo Methods
for Space Applications, ESA/ESTEC - Space Applications of Geant4 Polarized Processes,
G. Depaola and F. Longo, Round Table on Monte
Carlo Methods for Space Applications, ESA/ESTEC - Advanced software engineering in simulation
development and applications, M. G. Pia, Round
Table on Monte Carlo Methods for Space
Applications, ESA/ESTEC - Geant4 application to X and Gamma-ray Telescopes,
F. Longo et al., Round Table on Monte Carlo
Methods for Space Applications, ESA/ESTEC - Geant4-DNA, M. G. Pia, Round Table on Monte Carlo
Methods for Space Applications, ESA/ESTEC - Comparison between Plato isodose distribution of
a 192Ir source and those simulated with Geant4
toolkit, F. Foppiano, ESTRO
Congress, Stresa - Applicazione del toolkit Montecarlo Geant4 per la
dosimetria di trattamenti brachiterapici,
F.
Foppiano, AIFM Congress, Brescia
- From HEP computing to bio-medical research and
vice-versa technology transfer and application
results,
M. G. Pia et al., Plenary talk at CHEP 2001 - Architecture of Collaborating Frameworks,
A. Pfeiffer et al., CHEP2001 - Simulation For Astroparticle Experiments And
Planetary Explorations,
A. Brunengo et al., CHEP2001 - Geant4 Low Energy Electromagnetic Physics, M.
G. Pia et al., CHEP2001 - Simulation in the bio-medical domain
applications and results, M. G. Pia et al.,
International Conference on Advanced Technologies
for Particle Physics, Como - The Geant4 Toolkit simulation capabilities and
application results,
M. G. Pia, EPS-HEP Conference,
Budapest - The GEANT4 Simulation Toolkit, G. Santin, Monte
Carlo Workshop for Nuclear Medicine Applications,
Paris - Overview of Geant4, M. G. Pia, Round Table on
Monte Carlo Methods for Space Applications,
ESA/ESTEC
32Projects for 2002
- LowE OOAD, physics, testing etc. (see next
slide) - Hadronic physics tests and application examples
- Tracking further testing, test beam at Fermilab
- Advanced examples (user support testing)
- Underground background
- Hadronic physics?
- IORT
- Analysis in Geant4 (collaboration with Anaphe)
- Response to internal and external use cases
- Unit and regression testing
- General framework, collaboration with Anaphe
- Explore Aspect Oriented Programming, Subject
Oriented Programming - Software process
- Model based on RUP adapt to HEP (1st HEP
prototype, as far as we know) - Outreach (collaboration with CERN)
- Brochure, article on CERN Courier, seminar
33LowE e.m. activities proposed for 2002
- Physics
- Series of polarised LowE processes
- Alternative models (Compton Doppler broadening,
Photoelectric angular distribution, Penelope
models) - Positrons?
- PIXE
- Design iteration in e.m. physics
- Final state
- Energy loss
- Physics processes
- Testing
- Continue the validation of existing
implementations - Generalisation and automatisation of testing
(unit, regression, physics) - Statistical system testing
34Any interest in training?
- Introduction to Geant4 (seminar)
- Geant4 training courses (based on Geant4 Training
Kit) - Basic introduction
- Hands-on course
- Lectures (Villa Gualino, Torino)
- Introduction to OO methods
- Introduction to OOAD and related CASE tools
(Rose) - Introduction to advanced software engineering
methods - Software process
- SPI
- Valuable expertise available in INFN