Speaker
Abstract
We report about our experience regarding the porting of High Performance
Computing (HPC) applications MPI based to the GRID. In particular, we
ported FLASH, a multi-dimensional, adaptive-mesh, parallel code capable
of handling general flow problems in astrophysical environments. The
HPC simulations performed using FLASH require a substantial amount of
computational resources and intense and sustained communication activity
among the processes.
We describe here the features added to the Grid Infrastructure of the
COMETA Consortium in order to fulfill satisfactorily the requests of
HPC applications: the adoption of a fast interconnection network with low
communication latency; the gLite middleware extended to support MPI/MPI2
jobs; the newly developed license server and the specific scheduling
policy adopted; the use of watchdog utility for job monitoring during
execution; the long term proxy to allow the running of jobs whose
execution is particularly time-consuming and requiring hundreds of
processors.
