학술논문

The CMS Simulation Software
Document Type
Conference
Source
2006 IEEE Nuclear Science Symposium Conference Record Nuclear Science Symposium Conference Record, 2006. IEEE. 3:1655-1659 Oct, 2006
Subject
Nuclear Engineering
Power, Energy and Industry Applications
Fields, Waves and Electromagnetics
Engineered Materials, Dielectrics and Plasmas
Collision mitigation
Discrete event simulation
Detectors
Software performance
Event detection
Physics
Solid modeling
Data models
Electromagnetic fields
Particle tracking
Language
ISSN
1082-3654
Abstract
In this paper we present the features and the expected performance of the re-designed CMS simulation software, as well as the experience from the migration process. Today, the CMS simulation simulation suite is based on the two principal components - Geant4 detector simulation toolkit and the new CMS offline Framework and Event Data Model. The simulation chain includes event generation, detector simulation, and digitization steps. With Geant4, we employ the full set of electromagnetic and hadronic physics processes and detaild particle tracking in the 4 Tesla magnetic field. The Framework provides "action on demand" mechanisms, to allow users to load dynamically the desired modules and to configure and tune the final application at the run time. The simulation suite is used to model the complete central CMS detector (over 1 million of geometrical volumes) and the forward systems, such as Castor calorimeter and Zero Degree Calorimeter, the Totem telescopes, Roman Pots, and the Luminosity Monitor. The designs also previews the use of the electromagnetic and hadronic showers parametrization, instead of full modelling of high energy patricles passage through a complex heirarchy of volumes and materials, allowing significant gain in speed while tuning the simulation to test beam and collider data. Physics simulation has been extensively validated by comparison with test beam data and previous simulation results. The redesigned and upgraded simulation software was exercised for performance and robustness tests.