Quasi-Monte Carlo integration on the grid for sensitivity studies

In this paper we present error and performance analysis of quasi-Monte Carlo algorithms for solving multidimensional integrals (up to 100 dimensions) on the grid using MPI. We take into account the fact that the Grid is a potentially heterogeneous computing environment, where the user does not know the specifics of the target architecture. Therefore parallel algorithms should be able to adapt to this heterogeneity, providing automated load-balancing. Monte Carlo algorithms can be tailored to such environments, provided parallel pseudorandom number generators are available. The use of quasi-Monte Carlo algorithms poses more difficulties. In both cases the efficient implementation of the algorithms depends on the functionality of the corresponding packages for generating pseudorandom or quasirandom numbers. We propose efficient parallel implementation of the Sobol sequence for a grid environment and we demonstrate numerical experiments on a heterogeneous grid. To achieve high parallel efficiency we use a newly developed special grid service called Job Track Service which provides efficient management of available computing resources through reservations.

[1]  Kaizar Amin,et al.  QoS support for high-performance scientific Grid applications , 2004, IEEE International Symposium on Cluster Computing and the Grid, 2004. CCGrid 2004..

[2]  Ian T. Foster,et al.  Grid information services for distributed resource sharing , 2001, Proceedings 10th IEEE International Symposium on High Performance Distributed Computing.

[3]  Bob Jones,et al.  The Organization and Management of Grid Infrastructures , 2009, Computer.

[4]  Emanouil I. Atanassov,et al.  A Superconvergent Monte Carlo Method for Multiple Integrals on the Grid , 2005, International Conference on Computational Science.

[5]  Hongmei Chi,et al.  Efficient Generation of Parallel Quasirandom Faure Sequences Via Scrambling , 2007, International Conference on Computational Science.

[6]  Andreas Uhl,et al.  Techniques for parallel quasi-Monte Carlo integration with digital sequences and associated problems , 2001 .

[7]  Hongmei Chi,et al.  Generating parallel quasirandom sequences via randomization , 2007, J. Parallel Distributed Comput..

[8]  Marco Ratto,et al.  Global Sensitivity Analysis , 2008 .

[9]  Stefano Tarantola,et al.  Sensitivity Analysis in Practice: A Guide to Assessing Scientific Models , 2004 .

[10]  Fotis Georgatos,et al.  A Report on the Effect of Heterogeneity of the Grid Environment on a Grid Job , 2009, LSSC.

[11]  Harvey M. Wagner,et al.  Global Sensitivity Analysis , 1995, Oper. Res..

[12]  Ian Foster,et al.  The Grid 2 - Blueprint for a New Computing Infrastructure, Second Edition , 1998, The Grid 2, 2nd Edition.

[13]  Zahari Zlatev,et al.  Parallel and GRID Implementation of a Large Scale Air Pollution Model , 2006, Numerical Methods and Applications.

[14]  A. Owen.  THE DIMENSION DISTRIBUTION AND QUADRATURE TEST FUNCTIONS , 2003 .

[15]  Bruno Tuffin,et al.  A central limit theorem and improved error bounds for a hybrid-Monte Carlo sequence with applications in computational finance , 2006, J. Complex..

[16]  B. C. Bromley,et al.  Quasirandom Number Generators for Parallel Monte Carlo Algorithms , 1996, J. Parallel Distributed Comput..

[17]  John Soldatos,et al.  Programmable Grids Framework Enabling QoS in an OGSA Context , 2004, International Conference on Computational Science.

[18]  Anastas Mishev,et al.  Performance Analysis of GRID Middleware Using Process Mining , 2008, ICCS.

[19]  Anastas Mishev,et al.  User Level Grid Quality of Service , 2009, LSSC.

[20]  Ami Marowka,et al.  The GRID: Blueprint for a New Computing Infrastructure , 2000, Parallel Distributed Comput. Pract..

[21]  R. Caflisch.  Monte Carlo and quasi-Monte Carlo methods , 1998, Acta Numerica.

[22]  Emanouil I. Atanassov.  A New Efficient Algorithm for Generating the Scrambled Sobol' Sequence , 2002, Numerical Methods and Application.

[23]  Henri Casanova,et al.  Measuring the Performance and Reliability of Production Computational Grids , 2006, 2006 7th IEEE/ACM International Conference on Grid Computing.