With MPI jobs using an Isilon Storage system for scratch, are there I/o issues that can be avoided?

storage
isilon
hpc-arch-and-perf

#1

When running a storage intensive MPI job that uses an Isilon Storage system for scratch (working storage), are there any I/O bottlenecks that can be avoided?

Are there any clear ways of identifying them when they happen?

CURATOR: jpessin1