Implementation of an object oriented track reconstruction model into multiple LHC experiments

An Object Oriented (OO) model (Gaines et al., 1996; 1997; Gaines and Qian, 1998; 1999) for track reconstruction by the Kalman filtering method has bee n designed for high energy physics experiments at high luminosity hadron co lliders. The model has been coded in the C++ programming language and has b een successfully implemented into the OO computing environments of both the CMS (1994) and ATLAS (1994) experiments at the future Large Hadron Collide r (LHC) at CERN. We shall report: (1) how the OO model was adapted, with largely the same code, to different scenarios and serves the different reconstruction aims in different experim ents (i.e. the level-2 trigger software for ATLAS and the offline software for CMS); (2) how the OO model has been incorporated into different OO environments w ith a similar integration structure (demonstrating the ease of re-use of OO program); (3) what are the OO model's performance, including execution time, memory u sage, track finding efficiency and ghost rate, etc.; and (4) additional physics performance based on use of the OO tracking model. We shall also mention the experience and lessons learned from the implement ation of the OO model into the general OO software framework of the experim ents. In summary, our practice shows that the OO technology really makes th e software development and the integration issues straightforward and conve nient; this may be particularly beneficial for the general noncomputer-prof essional physicists. (C) 2001 Elsevier Science B.V. All rights reserved.