QosCosGrid e-Science Infrastructure for Large-Scale Complex System Simulations

This chapter focuses on two enhanced and widely used parallel computing environments: QosCosGrid-ProActive (QCG-ProActive) and QosCosGrid-OpenMPI (QCG-OMPI). QCG-OMPI is designed to enable parallel applications to run across geographically widely distributed computational resources owned and secured by different administrative organizations. QCG-ProActive was successfully integrated with the Repast Suite, a very popular Java-based, agent-based modeling and simulation platform. The chapter presents a short overview of validation scenarios. It discusses the main requirements of complex systems and other demanding parallel applications and classifies these into ?> templates. ?> The chapter introduces the QCG middleware services and their key capabilities relevant to end users. A number of useful libraries and frameworks for application developers and resource providers using QCG are described. The chapter provides additional Web-based monitoring and troubleshooting tools for QCG, which are already available on some production-level computational clusters and supercomputers. large-scale systems; middleware; modelling

[1]  Roger D. Hersch,et al.  Dynamic load balancing of parallel cellular automata , 2000, SPIE Optics + Photonics.

[2]  Matthias Scheutz,et al.  Adaptive algorithms for the dynamic distribution and parallel execution of agent-based models , 2006, J. Parallel Distributed Comput..

[3]  Jarek Nabrzyski,et al.  Dynamic grid scheduling with job migration and rescheduling in the GridLab resource management system , 2004, Sci. Program..

[4]  Denis Caromel,et al.  Programming, Composing, Deploying for the Grid , 2006, Grid Computing: Software Environments and Tools.

[5]  Thomas Hérault,et al.  QR factorization of tall and skinny matrices in a grid computing environment , 2009, 2010 IEEE International Symposium on Parallel & Distributed Processing (IPDPS).

[6]  László Gulyás,et al.  A Modeler-Friendly API for ABM Partitioning , 2009 .

[7]  Peter M. A. Sloot,et al.  Multi-scale modelling in computational biomedicine , 2010, Briefings Bioinform..

[8]  Andreas Haas,et al.  Standardization of an API for Distributed Resource Management Systems , 2007, Seventh IEEE International Symposium on Cluster Computing and the Grid (CCGrid '07).

[9]  Thomas Hérault,et al.  QCG-OMPI: MPI applications on grids , 2011, Future Gener. Comput. Syst..