Managing Parallel And Distributed Monte Carlo Simulations For Computational Finance In A Grid Environment

Computing in financial services consists of a combination of time-critical computations completed during trading hours, such as Monte Carlo simulations for option pricing, and over-night calculations on massive data sets, such as those required for market risk measurement. To date, this has typically been done using traditional parallel or cluster computing techniques. The French National Research Agency (ANR), along with several banks and financial software companies have partnered with INRIA to explore the application of grid computing to this domain. The PicsouGrid project utilizes the ProActive Java distributed computing library to parallelize and distribute Monte Carlo option pricing simulations, concurrently utilizing 102-103 workers. PicsouGrid has been deployed on various grid systems to evaluate its scalability and performance. Issues arising from the heterogeneity and layering of grid infrastructures are addressed via an abstract process model which is applied at each layer. Timings of both the algorithms and the grid infrastructures are carefully measured to provide better insight into the behavior and utilization of computational grids for this important class of parallel simulations.