Parallel program development and execution in the grid

Abstract

Integration of P-GRADE with Condor and parallel check-pointing results in an environment under which both interactive parallel program development and batch mode execution is possible for the Grid. The integrated P-GRADE/Condor Grid system will guarantee reliable, fault-tolerant parallel program execution in the Grid, like the Condor system guarantees such features for sequential programs. The GRM/PROVE performance monitoring and visualisation toolset has been separated from P-GRADE and extended towards the Grid to create the basis of a general Grid application monitoring infrastructure for parallel programs.