Communication Characteristics and Hybrid MPI/OpenMP Parallel Pr ogramming on Clusters of Multi-core SMP Nodes
作者： Rolf Rabenseifner , Georg Hager , Gabriele Jost
DOI:
关键词:
摘要: Hybrid MPI/OpenMP and pure MPI on clusters of multicore SMP nodes involve several mismatch problems between the parallel programming models hardware architectures. Measurements communication characteristics cores same socket, node, platforms (including Cray XT4 XT5) show that machine topology has a significant impact performance for all parallelization strategies awareness should be built into applications in future. We describe potentials challenges dominant hierarchically structured hardware. Case studies with multizone NAS benchmarks demonstrate opportunities hybrid programming.
cug.org
hlrs.de 参考文章(12)
Jin Haopiang, Rob F. vanderWijngaart, NAS Parallel Benchmarks, Multi-Zone Versions ,(2003)
Georg Hager, Holger Stengel, Thomas Zeiser, Gerhard Wellein, RZBENCH: Performance Evaluation of Current HPC Architectures Using Low-Level and Application Benchmarks High Performance Computing in Science and Engineering, Garching/Munich 2007. pp. 485- 501 ,(2009) , 10.1007/978-3-540-69182-2_39
Barbara M. Chapman, Lei Huang, Haoqiang Jin, Gabriele Jost, Bronis R. de Supinski, Toward Enhancing OpenMP’s Work-Sharing Directives Euro-Par 2006 Parallel Processing. pp. 645- 654 ,(2006) , 10.1007/11823285_67
Rolf Rabenseifner, Georg Hager, Gabriele Jost, Rainer Keller, Hybrid MPI and OpenMP Parallel Programming Recent Advances in Parallel Virtual Machine and Message Passing Interface. pp. 11- 11 ,(2006) , 10.1007/11846802_10
Matthew Curtis-Maury, Ankur Shah, Filip Blagojevic, Dimitrios S. Nikolopoulos, Bronis R. de Supinski, Martin Schulz, Prediction models for multi-dimensional power-performance optimization on many cores Proceedings of the 17th international conference on Parallel architectures and compilation techniques - PACT '08. pp. 250- 259 ,(2008) , 10.1145/1454115.1454151
Rolf Rabenseifner, Gerhard Wellein, Communication and Optimization Aspects of Parallel Programming Models on Hybrid Architectures ieee international conference on high performance computing data and analytics. ,vol. 17, pp. 49- 62 ,(2003) , 10.1177/1094342003017001005
Richard D. Loft, Stephen J. Thomas, John M. Dennis, Terascale spectral element dynamical core for atmospheric general circulation models Proceedings of the 2001 ACM/IEEE conference on Supercomputing (CDROM) - Supercomputing '01. pp. 18- 18 ,(2001) , 10.1145/582034.582052
C. Amza, A.L. Cox, S. Dwarkadas, P. Keleher, Honghui Lu, R. Rajamony, Weimin Yu, W. Zwaenepoel, TreadMarks: shared memory computing on networks of workstations IEEE Computer. ,vol. 29, pp. 18- 28 ,(1996) , 10.1109/2.485843
Rolf Rabenseifner, Georg Hager, Gabriele Jost, Hybrid MPI/OpenMP Parallel Programming on Clusters of Multi-Core SMP Nodes parallel, distributed and network-based processing. pp. 427- 436 ,(2009) , 10.1109/PDP.2009.43
Brice Goglin, High Throughput Intra-Node MPI Communication with Open-MX parallel, distributed and network-based processing. pp. 173- 180 ,(2009) , 10.1109/PDP.2009.20