SLAC PEP-II
BABAR
SLAC<->RAL
Babar logo
HEPIC E,S & H Databases PDG HEP preprints
Organization Detector Computing Physics Documentation
Personnel Glossary Sitemap Search Hypernews
Unwrap page!
Simulation Home
Sim Codes
Event Generators
Bogus/BgsApp
SimApp
Bear
Moose
Fast Simulation
Geant4 Home
Subsystems
PEP
SVT
DCH
DRC
EMC
IFR
Mixing/Trigger
Backgrounds
Mixing
Trigger Simulation
MC Truth/QA
MC Truth
Micro/Mini
QA Histograms
Sim Error Reports
REMEDY
MC Production
Production Home
Test Production
Tools
Database
CERNLIB
CLHEP
Event display
RandControl
Scripts
Check this page for HTML 4.01 Transitional compliance with the
W3C Validator
(More checks...)
BaBar   Computing BaBar Simulation Home
Getting Involved in Simulation
  Maintained by TW

BBSIM - Geant4 - Requirements - QA/QC - Aslund - Generators - Production
Tasks requiring subsystem-level participation

Many current areas of simulation development activity are listed below. A summary of areas in which involvement is needed from each detector subsystem is at the bottom. To get involved in any aspect of simulation, get in touch with me right away. We need you!
- Torre wenaus@llnl.gov

BBSIM detailed simulation based on Geant 3.21

BBSIM will be BaBar's production simulation until the Geant4-based simulation is complete and fully validated. New BBSIM development is done in such a way as to be usable in the context of Geant4 as well. We have developed an approach allowing virtually all BBSIM code to be written either in Fortran or in C++ (already the geometry can be in either language), in the latter case directed towards later reuse with Geant4. BBSIM work doesn't necessarily mean Fortran coding; it can provide a means of learning the basics of C++ while doing something productive.
  • Development of detailed geometrical and material models of the detector subsystems using Geant 3.21. Detailed models of all subsystems exist but most are being continually refined. Subsystem coordinators have the current to-do list.
  • Refinement of detector response models (eg. drift-time relation and energy loss in the DCH, Cerenkov propagation in the DIRC, nonuniform light collection in the calorimeter). Subsystem coordinators will have the current to-do list.
  • Modelling of the signal collection and digitization processes in the subsystems
  • Design of the data structures, data flow, program modularization and control through the different simulation steps from geometry initialization and event generation to tracking, hit generation and digitization, and on to reconstruction. This is largely completed.
  • Incorporation of needed calibration, alignment, and configuration information into the simulation; interaction with the BaBar database

Final simulation based on Geant4 (eventual BBSIM and Aslund replacement)

BaBar's Geant4 simulation will support fast as well as detailed simulation, with variable levels of detail selectable at the subsystem level. Tasks relating to fast and full simulation in the Geant4 context are described here.
  • Direct participation in design and development of the object oriented simulation toolkit Geant4, Geant321's successor. BaBar participation focuses on subprojects essential to our needs (eg. fast parameterisation functionality, managing hits and other output objects and their interface to BaBar codes, managing the front-end generator input and propagation of generator info out to downstream codes, ...) to ensure our requirements are met and are delivered on our schedule.
  • Design and development of the BaBar application layer over Geant4, eg. a framework supporting simulation at variable levels of detail (fast to full detail), geometry and parameterised simu definition via dbio/OODB, Geant4 integration into SRT/Framework. The Bogus prototype is the basis for this work.
  • Participation in G4-based simulation implementation and validation by migrating existing simulations to Geant4 using migration tools (g3tog4, euclid) for full simulation and adapting Aslund parameterisations, and cross-checking G4 performance against BBSIM.
  • Creation of a dual-language development environment for BBSIM supporting code reuse in Geant4. Tools in place; need to be documented with examples.
  • Writing reusable C++ BBSIM code for later reuse with Geant4
  • Monitoring of and interfacing to associated development projects such as object database data storage and visualization
  • Interfacing the simulation to CAD engineering models. Geant4 accepts STEP-format CAD input allowing us to cross-check our simulation against the engineering model, eg. to check material traversed.
  • External collaboration (eg. fast tracking and calorimeter parameterisation work of Fermilab's MCFast project)
  • Interfaces: online/simulation (raw data), generator/simulation, database/simulation, production system
  • Simulation physics issues, such as improvements needed in hadronic physics; interaction with Geant4 and experts in and outside BaBar. Implementation of interaction physics/response model based on test beam results

Simulation Requirements Documentation

A priority task is formalization of simulation requirements (detail levels required by the subsystems and physics analysis, detector response and digitization, alignment and calibration, etc.) in requirements documents. These documents will provide the guidance to yield a final simulation that optimally balances realism against resource usage in the various application domains of detector simulation.

Quality Assurance and Quality Control (QA/QC)

QA/QC is the weakest part of the simulation effort today. BaBar is assembling a dedicated QA/QC group but we need to assemble a working group to prioritize and develop this area on a shorter timescale than the formation of the dedicated group, and to interface to the QA/QC group in the longer term. Some aspects of QA/QC:
  • Functionality test suite to validate releases
  • Performance validation across releases
  • Performance validation across platforms
  • Geant3-Geant4 comparison and Geant4 validation

Aslund fast simulation

The Aslund simulation is being maintained by the Orsay group. Its functionality is more or less frozen; new development work is going into the Geant4-based simulation.
  • Development and refinement of subsystem performance parametrizations based on BBSIM
  • Speed optimization

Physics Generators

Physics generator development is being coordinated by Walter Toki.
  • Collection, development, integration and support of needed physics generators, in collaboration with the physics group, and in the context of the BEGET/EvtGen event generator framework

Simulation production

  • The successor to the simulation production system developed for the 1996 productions and MDC1 is currently being developed by Paul Raines.
  • BaBar is still in need of a production manager.

Summary of tasks requiring subsystem-level participation:

  • QA/QC, test protocols
  • Requirements documentation
  • Geometry migration to Geant4, or development of new G4-based simulation with G3 port via euclid (for G3 cross-check)
  • Hit implementation in Geant4
  • Implementation of parameterised response models in G4
  • Incorporation of test beam data/results