Computational Infrastructure for Parallel, Distributed, and Grid-Based Monte Carlo Computations

摘要

Monte Carlo applications are widely perceived as computationally intensive but naturally parallel. Therefore, they can be effectively executed using the dynamic bag-of-work model which is well suited to parallel, distributed, and grid-based architectures. This paper concentrates on providing computational infrastructure for Monte Carlo applications on such architectures. This is accomplished by analyzing the characteristics of large-scale Monte Carlo computations, and leveraging the existing Scalable Parallel Random Number Generators (SPRNG) library. Based on these analyses, we improve the efficiency of subtask-scheduling by implementing and analyzing the "N-out-of-M" strategy, and develop a Monte Carlo-specific lightweight checkpointing technique, which leads to a performance improvement. Also, we enhance the trustworthiness of Monte Carlo applications on these architectures by utilizing the statistical nature of Monte Carlo and by cryptographically validating intermediate results utilizing the random number generator already in use in the Monte Carlo application. All these techniques lead to a high-performance grid-computing infrastructure that is capable of providing trustworthy Monte Carlo computation services.