New EuroCC-UK Technical Report: Increasing the MPI-IO performance on ARCHER2


File input and output can become bottlenecks for parallel programs running on large numbers of processors. EuroCC-UK staff at EPCC investigated how to increase the MPI-IO performance on ARCHER2, the […]

File input and output can become bottlenecks for parallel programs running on large numbers of processors. EuroCC-UK staff at EPCC investigated how to increase the MPI-IO performance on ARCHER2, the UK National HPC service, using the Cray Lustre lockahead options, which had not been previously investigated on this system. This was followed by an investigation of the performance of ADIOS2, a more modern parallel IO library. Finally, the team looked at the performance of IO from a real HPC application rather than a synthetic benchmark.

Read the report

DoI: 10.5281/zenodo.7643563


YOU MAY ALSO LIKE