On implementing MPI-IO portably and with high performance
Proceedings of the sixth workshop on I/O in parallel and distributed systems
Exploiting Transparent Remote Memory Access for Non-Contiguous- and One-Sided-Communication
IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
Data Sieving and Collective I/O in ROMIO
FRONTIERS '99 Proceedings of the The 7th Symposium on the Frontiers of Massively Parallel Computation
Fast Parallel Non-Contiguous File Access
Proceedings of the 2003 ACM/IEEE conference on Supercomputing
Transparent log-based data storage in MPI-IO applications
PVM/MPI'07 Proceedings of the 14th European conference on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Hi-index | 0.00 |
We explore several methods utilizing system-wide shared memory to improve the performance of MPI-IO, particularly for non-contiguous file access. We introduce an abstraction called the datatype iterator that permits efficient, dynamic generation of (offset, length) pairs for a given MPI derived datatype. Combining datatype iterators with overlapped I/O and computation, we demonstrate how a shared memory MPI implementation can utilize more than 90% of the available disk bandwidth (in some cases representing a 5× performance improvement over existing methods) even for extreme cases of non-contiguous datatypes. We generalize our results to suggest possible parallel I/O performance improvements on systems without global shared memory.