Parallelization of events generation for data analysis techniques

With the startup of the LHC experiments at CERN, the involved community is now focusing on the analysis of the collected data. The complexity of the data analyses will be a key factor for finding eventual new phenomena. For such a reason many data analysis tools have been developed in the last several years, which implement several data analysis techniques. Goal of these techniques is the possibility of discriminating events of interest and measuring parameters on a given input sample of events, which are themselves defined by several variables. Also particularly important is the possibility of repeating the determination of the parameters by applying the procedure on several simulated samples, which are generated using Monte Carlo techniques and the knowledge of the probability density functions of the input variables. This procedure achieves a better estimation of the results. Depending on the number of variables, complexity of their probability density functions, number of events, and number of sample to generate, the whole procedure can be high CPU-time consuming. In this paper we show how the Monte Carlo generation of the events for each simulated sample can be parallelized using OpenMP to scale over multi-cores in a single computational node.