Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters
The performance of MPI implementation operations still presents critical issues for high performance computing systems, particularly for more advanced processor technology. Consequently, this study concentrates on benchmarking MPI implementation on multi-core architecture by measuring the performanc...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Science Publications
2013
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/30612/ http://psasir.upm.edu.my/id/eprint/30612/1/Performance%20analysis%20of%20Message%20Passing%20Interface%20collective%20communication%20on%20intel%20xeon%20quad.pdf |
| _version_ | 1848846727528841216 |
|---|---|
| author | Ismail, Roswan Abdul Hamid, Nor Asilah Wati Othman, Mohamed Latip, Rohaya |
| author_facet | Ismail, Roswan Abdul Hamid, Nor Asilah Wati Othman, Mohamed Latip, Rohaya |
| author_sort | Ismail, Roswan |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | The performance of MPI implementation operations still presents critical issues for high performance computing systems, particularly for more advanced processor technology. Consequently, this study concentrates on benchmarking MPI implementation on multi-core architecture by measuring the performance of Open MPI collective communication on Intel Xeon dual quad-core Gigabit Ethernet and InfiniBand clusters using SKaMPI. It focuses on well known collective communication routines such as MPI-Bcast, MPI-AlltoAll, MPI-Scatter and MPI-Gather. From the collection of results, MPI collective communication on InfiniBand clusters had distinctly better performance in terms of latency and throughput. The analysis indicates that the algorithm used for collective communication performed very well for all message sizes except for MPI-Bcast and MPI-Alltoall operation of inter-node communication. However, InfiniBand provides the lowest latency for all operations since it provides applications with an easy to use messaging service, compared to Gigabit Ethernet, which still requests the operating system for access to one of the server communication resources with the complex dance between an application and a network.
|
| first_indexed | 2025-11-15T09:07:18Z |
| format | Article |
| id | upm-30612 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T09:07:18Z |
| publishDate | 2013 |
| publisher | Science Publications |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-306122015-09-21T02:12:13Z http://psasir.upm.edu.my/id/eprint/30612/ Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters Ismail, Roswan Abdul Hamid, Nor Asilah Wati Othman, Mohamed Latip, Rohaya The performance of MPI implementation operations still presents critical issues for high performance computing systems, particularly for more advanced processor technology. Consequently, this study concentrates on benchmarking MPI implementation on multi-core architecture by measuring the performance of Open MPI collective communication on Intel Xeon dual quad-core Gigabit Ethernet and InfiniBand clusters using SKaMPI. It focuses on well known collective communication routines such as MPI-Bcast, MPI-AlltoAll, MPI-Scatter and MPI-Gather. From the collection of results, MPI collective communication on InfiniBand clusters had distinctly better performance in terms of latency and throughput. The analysis indicates that the algorithm used for collective communication performed very well for all message sizes except for MPI-Bcast and MPI-Alltoall operation of inter-node communication. However, InfiniBand provides the lowest latency for all operations since it provides applications with an easy to use messaging service, compared to Gigabit Ethernet, which still requests the operating system for access to one of the server communication resources with the complex dance between an application and a network. Science Publications 2013-04 Article PeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/30612/1/Performance%20analysis%20of%20Message%20Passing%20Interface%20collective%20communication%20on%20intel%20xeon%20quad.pdf Ismail, Roswan and Abdul Hamid, Nor Asilah Wati and Othman, Mohamed and Latip, Rohaya (2013) Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters. Journal of Computer Science, 9 (4). 455-462. ISSN 1549-3636; ESSN: 1552-6607 http://thescipub.com/abstract/10.3844/jcssp.2013.455.462 10.3844/jcssp.2013.455.462 |
| spellingShingle | Ismail, Roswan Abdul Hamid, Nor Asilah Wati Othman, Mohamed Latip, Rohaya Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| title | Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| title_full | Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| title_fullStr | Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| title_full_unstemmed | Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| title_short | Performance analysis of Message Passing Interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| title_sort | performance analysis of message passing interface collective communication on intel xeon quad-core gigabit ethernet and infiniband clusters |
| url | http://psasir.upm.edu.my/id/eprint/30612/ http://psasir.upm.edu.my/id/eprint/30612/ http://psasir.upm.edu.my/id/eprint/30612/ http://psasir.upm.edu.my/id/eprint/30612/1/Performance%20analysis%20of%20Message%20Passing%20Interface%20collective%20communication%20on%20intel%20xeon%20quad.pdf |