Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers

Handle URI:
http://hdl.handle.net/10754/599158
Title:
Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers
Authors:
Wu, Xingfu; Taylor, Valerie
Abstract:
The NAS Parallel Benchmarks (NPB) are well-known applications with the fixed algorithms for evaluating parallel systems and tools. Multicore supercomputers provide a natural programming paradigm for hybrid programs, whereby OpenMP can be used with the data sharing with the multicores that comprise a node and MPI can be used with the communication between nodes. In this paper, we use SP and BT benchmarks of MPI NPB 3.3 as a basis for a comparative approach to implement hybrid MPI/OpenMP versions of SP and BT. In particular, we can compare the performance of the hybrid SP and BT with the MPI counterparts on large-scale multicore supercomputers. Our performance results indicate that the hybrid SP outperforms the MPI SP by up to 20.76%, and the hybrid BT outperforms the MPI BT by up to 8.58% on up to 10,000 cores on BlueGene/P at Argonne National Laboratory and Jaguar (Cray XT4/5) at Oak Ridge National Laboratory. We also use performance tools and MPI trace libraries available on these supercomputers to further investigate the performance characteristics of the hybrid SP and BT.
Citation:
Wu X, Taylor V (2011) Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers. SIGMETRICS Perform Eval Rev 38: 56. Available: http://dx.doi.org/10.1145/1964218.1964228.
Publisher:
Association for Computing Machinery (ACM)
Journal:
ACM SIGMETRICS Performance Evaluation Review
KAUST Grant Number:
KUS-I1-010-01
Issue Date:
29-Mar-2011
DOI:
10.1145/1964218.1964228
Type:
Article
ISSN:
0163-5999
Sponsors:
This work is supported by NSF grant CNS-0911023, and theAward No. KUS-I1-010-01 made by King Abdullah University ofScience and Technology (KAUST). The authors would like toacknowledge Argonne Leadership Computing Facility at ArgonneNational Laboratory for the use of BlueGene/P and National Centerfor Computational Science at Oak Ridge National Laboratory forthe use of Jaguar and JaguarPF under DOE INCITE project“Performance Evaluation and Analysis Consortium End Station”,and Haoqiang Jin from NASA Ames Research Center for providinghis BT code.
Appears in Collections:
Publications Acknowledging KAUST Support

Full metadata record

DC FieldValue Language
dc.contributor.authorWu, Xingfuen
dc.contributor.authorTaylor, Valerieen
dc.date.accessioned2016-02-25T13:53:59Zen
dc.date.available2016-02-25T13:53:59Zen
dc.date.issued2011-03-29en
dc.identifier.citationWu X, Taylor V (2011) Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers. SIGMETRICS Perform Eval Rev 38: 56. Available: http://dx.doi.org/10.1145/1964218.1964228.en
dc.identifier.issn0163-5999en
dc.identifier.doi10.1145/1964218.1964228en
dc.identifier.urihttp://hdl.handle.net/10754/599158en
dc.description.abstractThe NAS Parallel Benchmarks (NPB) are well-known applications with the fixed algorithms for evaluating parallel systems and tools. Multicore supercomputers provide a natural programming paradigm for hybrid programs, whereby OpenMP can be used with the data sharing with the multicores that comprise a node and MPI can be used with the communication between nodes. In this paper, we use SP and BT benchmarks of MPI NPB 3.3 as a basis for a comparative approach to implement hybrid MPI/OpenMP versions of SP and BT. In particular, we can compare the performance of the hybrid SP and BT with the MPI counterparts on large-scale multicore supercomputers. Our performance results indicate that the hybrid SP outperforms the MPI SP by up to 20.76%, and the hybrid BT outperforms the MPI BT by up to 8.58% on up to 10,000 cores on BlueGene/P at Argonne National Laboratory and Jaguar (Cray XT4/5) at Oak Ridge National Laboratory. We also use performance tools and MPI trace libraries available on these supercomputers to further investigate the performance characteristics of the hybrid SP and BT.en
dc.description.sponsorshipThis work is supported by NSF grant CNS-0911023, and theAward No. KUS-I1-010-01 made by King Abdullah University ofScience and Technology (KAUST). The authors would like toacknowledge Argonne Leadership Computing Facility at ArgonneNational Laboratory for the use of BlueGene/P and National Centerfor Computational Science at Oak Ridge National Laboratory forthe use of Jaguar and JaguarPF under DOE INCITE project“Performance Evaluation and Analysis Consortium End Station”,and Haoqiang Jin from NASA Ames Research Center for providinghis BT code.en
dc.publisherAssociation for Computing Machinery (ACM)en
dc.titlePerformance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputersen
dc.typeArticleen
dc.identifier.journalACM SIGMETRICS Performance Evaluation Reviewen
dc.contributor.institutionTexas A&M University, College Station, TXen
kaust.grant.numberKUS-I1-010-01en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.