Performance pattern of unified parallel C on multi-core clusters
The Partitioned Global Address Space (PGAS) model has been widely used in multi-core clusters as an alternative to MPI. Among the widespread use is Unified Parallel C (UPC). Previous research has shown that UPC performance is comparable with MPI, however in certain cases UPC require hand-tuning tech...
| Main Authors: | , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Published: |
IEEE (IEEE Xplore)
2013
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/41309/ |
| _version_ | 1848849661240016896 |
|---|---|
| author | Abdul Hamid, Nor Asilah Wati Serres, Olivier Anbar, Ahmad Hassan, Sazlinah |
| author_facet | Abdul Hamid, Nor Asilah Wati Serres, Olivier Anbar, Ahmad Hassan, Sazlinah |
| author_sort | Abdul Hamid, Nor Asilah Wati |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | The Partitioned Global Address Space (PGAS) model has been widely used in multi-core clusters as an alternative to MPI. Among the widespread use is Unified Parallel C (UPC). Previous research has shown that UPC performance is comparable with MPI, however in certain cases UPC require hand-tuning techniques such as prefetching and privatized pointers-to-shared to improve the performance. In this paper we reviews, evaluate and analyze the performance pattern between UPC Naïve, UPC optimize and MPI on two different multi-core clusters architecture. We focus our study using matrix multiplication as the benchmark and perform our experimental on two distributed memory machine, Cray XE6 with Gemini interconnects and Sun Cluster with Infiniband interconnects. We provide analysis on each core execution time to understand the pattern of communication for both machines. We also demonstrate the gaps between naïve and optimized are depends on the compiler with its associate distributed memory machine. We also observed unnecessary optimization for certain programs related to HPC architecture and compiler. |
| first_indexed | 2025-11-15T09:53:56Z |
| format | Conference or Workshop Item |
| id | upm-41309 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-15T09:53:56Z |
| publishDate | 2013 |
| publisher | IEEE (IEEE Xplore) |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-413092015-11-03T04:00:30Z http://psasir.upm.edu.my/id/eprint/41309/ Performance pattern of unified parallel C on multi-core clusters Abdul Hamid, Nor Asilah Wati Serres, Olivier Anbar, Ahmad Hassan, Sazlinah The Partitioned Global Address Space (PGAS) model has been widely used in multi-core clusters as an alternative to MPI. Among the widespread use is Unified Parallel C (UPC). Previous research has shown that UPC performance is comparable with MPI, however in certain cases UPC require hand-tuning techniques such as prefetching and privatized pointers-to-shared to improve the performance. In this paper we reviews, evaluate and analyze the performance pattern between UPC Naïve, UPC optimize and MPI on two different multi-core clusters architecture. We focus our study using matrix multiplication as the benchmark and perform our experimental on two distributed memory machine, Cray XE6 with Gemini interconnects and Sun Cluster with Infiniband interconnects. We provide analysis on each core execution time to understand the pattern of communication for both machines. We also demonstrate the gaps between naïve and optimized are depends on the compiler with its associate distributed memory machine. We also observed unnecessary optimization for certain programs related to HPC architecture and compiler. IEEE (IEEE Xplore) 2013 Conference or Workshop Item NonPeerReviewed Abdul Hamid, Nor Asilah Wati and Serres, Olivier and Anbar, Ahmad and Hassan, Sazlinah (2013) Performance pattern of unified parallel C on multi-core clusters. In: 2013 IEEE International Conference on High Performance Computing and Communications & 2013 IEEE International Conference on Embedded and Ubiquitous Computing, 13-15 Nov. 2013, Zhangzijie, China. (pp. 1751-1757). 10.1109/HPCC.and.EUC.2013.250 |
| spellingShingle | Abdul Hamid, Nor Asilah Wati Serres, Olivier Anbar, Ahmad Hassan, Sazlinah Performance pattern of unified parallel C on multi-core clusters |
| title | Performance pattern of unified parallel C on multi-core clusters |
| title_full | Performance pattern of unified parallel C on multi-core clusters |
| title_fullStr | Performance pattern of unified parallel C on multi-core clusters |
| title_full_unstemmed | Performance pattern of unified parallel C on multi-core clusters |
| title_short | Performance pattern of unified parallel C on multi-core clusters |
| title_sort | performance pattern of unified parallel c on multi-core clusters |
| url | http://psasir.upm.edu.my/id/eprint/41309/ http://psasir.upm.edu.my/id/eprint/41309/ |