-1679476342.png){kind=logo}
Abstract
Atmospheric models usually demand high processing power and generate large amounts of data. As the degree of parallelism grows, the I/O operations may become the major impacting factor of their performance. This work shows that a hybrid MPI/OpenMP implementation can improve the performance of the atmospheric model ocean-land-atmosphere model (OLAM) on a multicore cluster environment. We show that the hybrid MPI/OpenMP version of OLAM decreases the number of output files, resulting in better performance for I/O operations. We have evaluated OLAM on the parallel file system PVFS and shown that storing the files on PVFS results in lower performance than using the local disks of the cluster nodes due as a consequence of file creation and network concurrency. We have also shown that further parallel optimisations should be included in the hybrid version in order to improve the parallel execution time of OLAM.
Keywords
References
- 1. (2009).
‘Scalable I/O forwarding framework for high-performance computing systems’.
CLUSTER ‘09. IEEE International Conference on ,September 2009 ,New Orleans, Louisiana, USA , 1-10 Google Scholar - 2. (2011). ‘I/O performance of a large atmospheric model using PVFS’. in Rencontres francophones du Paralllisme (RenPar20). Google Scholar
- 3. (2009). ‘Small-file access in parallel file systems’. Parallel and Distributed Processing, IPDPS. 1-11 Google Scholar
- 4. (2000). ‘PVFS: a parallel file system for Linux clusters’. in Proceedings of the 4th conference on 4th Annual Linux Showcase and Conference (ALS00). USENIX Association Google Scholar
- 5. (2011). ‘Optimizing code Saturne computations on Petascale systems’. Journal of Computer & Fluids. 45, 1, 103-108 Google Scholar
- 6. (2008). ‘WRF nature run’. Journal of Physics: Conference Series. 125, 1, 012022 Google Scholar
- 7. (2009). ‘Improving parallel write by node-level request scheduling’. in CCGRID 09: Proceedings of the 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid. , 203196- Google Scholar
- 8. (2010). ‘I/O performance evaluation on multicore clusters with atmospheric model environment’. Proceedings of 22nd International Symposium on Computer Architecture and High Performance Computing-WAMMCA. Google Scholar
- 9. Partnership for Advanced Computing in Europe (PRACE) available at http://www.prace-project.eu Google Scholar
- 10. (1992). ‘A comprehensive meteorological modeling system – RAMS’. Meteorology and Atmospheric Physics. 49, 1, 69-91 Google Scholar
- 11. (2010). ‘Performance evaluation of an atmospheric simulation model on multi-core environments’. Proceedings of the Latin American Conference on High Performance Computing (CLCAR). 330-332 Google Scholar
- 12. (2009).
‘Application level I/O caching on blue gene/P systems’.
IEEE International Symposium on Parallel & Distributed Processing, IPDPS 2009 ,Rome, Italy , 1-8 Google Scholar - 13. ‘OLAM: ocean-land-atmosphere model – model input parameters – Version 3.0’. 2008, 11, Duke University, Tech. Rep. Google Scholar
- 14. (2012). ‘Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore clusters’. in The Computer Journal. 55, 2, 154167 Google Scholar
- 15. (2011). ‘Performance-based parallel loop self-scheduling using hybrid OpenMP and MPI programming on multicore SMP clusters’. in Concurrency and Computation: Practice and Experience. 721-744 Google Scholar
