The Florida State University College of Arts and Sciences a Grid Computing Infrastructure for Monte Carlo Applications

Monte Carlo applications are widely perceived as computationally intensive but naturally parallel. Therefore, they can be effectively executed on the grid using the dynamic bag-of-work model. We improve the efficiency of the subtask-scheduling scheme by using an N-out-of-M strategy, and develop a Monte Carlo-specific lightweight checkpoint technique, which leads to a performance improvement for Monte Carlo grid computing. Also, we enhance the trustworthiness of Monte Carlo grid-computing applications 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 our implementation of a gridcomputing infrastructure – GCIMCA (Grid-Computing Infrastructure for Monte Carlo applications), which is based on Globus and the SPRNG (Scalable Parallel Random Number Generators) library. GCIMCA intends to provide trustworthy grid-computing services for large-scale and high-performance distributed Monte Carlo computations. We apply Monte Carlo applications to GCIMCA to show the capability of our techniques. These applications include the grid-based Monte Carlo integration and a “real-life” Monte Carlo application -the grid-based hybrid Molecular Dynamics (MD)/Brownian Dynamics (BD) application for simulating the long-time, nonequilibrium dynamics of receptor-ligand interactions. Our preliminary results show that our techniques and infrastructure can achieve significant speedup, efficiency, accuracy, and trustworthiness for grid-based Monte Carlo applications.