GEANT 4 for Future Linear Colliders Geant 4

GEANT 4 for Future Linear Colliders Geant 4 Workshop @ DESY October 2, 2003

Linear Collider Environment n Exploit the physics discovery potential of e+ecollisions at s ~ 1 Te. V. n Precision measurements of complex final states require detectors with: Exceptional momentum resolution & vertexing. n Imaging calorimetry for “Energy Flow” analysis. n n Common simulation environment for all LC studies would allow sharing of detectors, algorithms, and code. n The system should be flexible, powerful, yet simple to install and maintain.

LC Detector Full Simulation MC Event G 4 Application Raw Event Geometry GEANT 4 Geometry Database Reconstruction, Visualization, …

Mokka n Geant 4 full simulation for the Tesla detector. n Uses subdetector-specific geometry drivers. Relevant parameters stored in My. SQL database. n Tight coupling between Sensitive Detector and geometry volume definitions. n n LCIO persistence for generic hits & MC chain.

The Proto 00 geometry driver P 55 MOKKA Proto 00 P 55 Ec

LCD Full Simulation n Geometry defined in XML. Flexible, but simplified volumes. n Projective readout of sensitive volumes. n n Dynamic topology, not just parameters. n Have defined generic hit classes for sensitive tracker and calorimeter hits. n Root and LCIO bindings for I/O.

TPC Tracker, Si Disks, CCD VTX

All Si Tracker, CCD VTX

Generic Hits Problem Statement n We wish to define a generic output hit format for full simulations of the response of detector elements to physics events. n Want to preserve the “true” Monte Carlo track information for later comparisons. n Want to defer digitization as much as possible to allow various resolutions, etc. to be efficiently studied.

Types of Hits n “Tracker” Hits Position sensitive. n Particle unperturbed by measurement. n Save “ideal” hit information. n n “Calorimeter” Hits Energy sensitive. n Enormous number of particles in shower precludes saving of each “ideal” hit. n Quantization necessary at simulation level. n

Hits Summary n Storing “ideal” hits gives detailed information about MC track trajectory. n Deferring digitization allows studies of detector resolution to be efficiently conducted. n Can approximate the same in calorimeter by defining small cells, then ganging later.

LCIO n Persistency framework for LC simulations. n Currently uses SIO: Simple Input Output on the fly data compression n some OO capabilities, e. g. pointers n C++ and Java implementation available n n Changes in IO engine designed for. n Extensible event data model Generic Tracker and Calorimeter Hits. n Monte Carlo particle heirarchy. n

LCIO (II) n Persistency framework for LC simulations. n Java, C++ and f 77 user interface. n LCIO is currently implemented in simulation frameworks: hep. lcd n Mokka/BRAHMS-reco n -> other groups are invited to join

LCIO Motivation LCIO Persistency Framework Generator Java, C++, Fortran Geant 3, Geant 4 Simulation Java, C++, Fortran Reconstruction geometry Java, C++, Fortran Analysis

Towards Internationalization n Suggest that Tesla, NLC and JLC full simulation groups could run a single GEANT 4 executable. n Geometry determined at run-time (XML). n Write out common “ideal” hits. n Digitize as appropriate with plug-ins. n Enormous savings in effort. n Makes comparisons easy.

Full Simulations LCD Full Sim GISMO C++ LCDROOT/LCDG 4 BRAHMS JIM GEANT 3 FORTRAN MOKKA Common GEANT 4 executable Runtime geometry Generic Hit output JUPITER

LCD/Mokka n First version of mysql / xml interface exists n SD detector fully modelled including beamline elements. n Several TESLA detector versions modelled. n LCIO output implemented in beta version. n Interfaces to HEPEVT and STDHEP and background files implemented. n Interface to AIDA integrated.

SD in Mokka

LC Detector Full Simulation MC Event (STDHEP) Histograms (AIDA) G 4 Application Geometry (XML) Database (My. SQL) (Generic Hits) (LCIO/Root) GEANT 4 Geometry Raw Event

Main Simulation Issues n Need n flexible method to describe geometry. Prefer G 4 supported geometry input (GDML? ) n Beam Delivery System requires arbitrary magnetic fields, excellent tracking precision. -5 Ge. V/c n Tracking System: (1/p. T) 5 x 10 (1/p. T) n Multiple Scattering, tracking precision. n Jet n Reconstruction: E/E~30%/√E Excellent hadronic shower simulations.

Highlights of LC Geant 4 Effort n Common n executable, with runtime geometry. Detector designs compared on equal footing. n Generic hits for trackers and calorimeters. Simplifies Sensitive Detector implementation. n Post-GEANT digitization design flexibility. n n Lightweight persistence format (LCIO). Allows interchange of data between communities. n Common target for Java, C++ & Fortran analyses. n

Why XML? n Simplicity: Rigid set of rules, plain text n Extensibility: Add custom features, data types n Interoperability: between OS and languages n Self-describing data n Hierarchical structure OOP n Open W 3 standard, lingua franca for B 2 B n Many tools for validating, parsing, translating n Automatic code-generation for data-binding

Why G 4 XML? n XML Schema very useful for “compile-time” type safety and bounds checking. n Prefer a G 4 -supported XML-based solution. Had hoped for common LHC solution. n Investigated GDML. n Looks promising. n Sensitive detector definitions needed. n Support? n