ABSTRACT
XcalableMP is a parallel extension of existing languages, such as C and Fortran, that was proposed as a new programming model to facilitate program parallel applications for distributed memory systems. In order to investigate the performance of parallel programs written in XcalableMP, we have implemented NAS Parallel Benchmarks, specifically, the Embarrassingly Parallel (EP) benchmark, the Integer Sort (IS) benchmark, and the Conjugate Gradient (CG) benchmark, using XcalableMP. The results show that the performance of XcalableMP is comparable to that of MPI. In particular, the performances of IS with a histogram and CG with two-dimensional parallelization achieve almost the same performance. The results also demonstrate that XcalableMP allows a programmer to write efficient parallel applications at a lower programming cost.
- Message Passing Interface Forum. http://www.mpi-forum.org/Google Scholar
- Directive-based language eXtension for Scalable and performance-aware Parallel Programming. http://www.xcalablemp.org/Google Scholar
- Jinpil Lee and Mitsuhisa Sato, Implementation and Performance Evaluation of XcalableMP: A Parallel Programming Language for Distributed Memory Systems, 39th Annual International Conference on Parallel Processing (2010) Google ScholarDigital Library
- http://www.xcalablemp.org/HPCC09_XMP.pdfGoogle Scholar
- OpenMP http://openmp.org/wp/Google Scholar
- Robert W. Numrich and John Reid. Co-array Fortran for parallel programming, ACM SIGPLAN Fortran Forum Volume 17 Issue 2 (1998) Google ScholarDigital Library
- JAHPF (Japan Association of High Performance Fortran). http://www.hpfpc.org/jahpf/Google Scholar
- Yuanyuan Zhang, Hidetoshi Iwashita, Kuninori Ishii, Masanori Kaneko, Tomotake Nakamura and Kohichiro Hotta Hybrid Parallel Programming on SMP Clusters Using XPFortran and OpenMP, 6th International Workshop on OpenMP (IWOMP 2010), LNCS 6132, pp. 133--148, Springer, Heidelberg (2010) Google ScholarDigital Library
- Iwashita, H., Sueyasu, N., Kamiya, S., van Waveren, M. VPP Fortran and the Design of HPF/JA Extensions. In: Concurrency and Computation: Practice and Experience, vol. 14(8-9), pp. 575--588. John Wiley & Sons Ltd., Chichester (2002)Google Scholar
- Jinpil Lee, Mitsuhisa Sato and Taisuke Boku. OpenMPD: A Directive Based Data Parallel Language Extensions for Distributed Memory Systems, Proceedings of the 37th International Conference on Parallel Processing, pp. 121--128, 2008 Google ScholarDigital Library
- http://www.hpcchallenge.org/Google Scholar
- Bailey, D. H. and et al. THE NAS PARALLEL BENCHMARKS, Technical Report NAS-94-007, Nasa Ames Research Center (1994)Google Scholar
- Frumkin, M., Jin, H. and Yan, J.: Implementation of NAS parallel benchmarks in high peroformance fortran, Technical Report NAS-98-009, Nasa Ames Research Center (1998)Google Scholar
- Nishitani, Y. et al.: Techiniques for compiling and implementing all NAS parallel benchmarks in HPF, Concurrency and Computation - Practice & Experience, Vol. 14, No. 8-9, Wiley, pp. 769--787 (2002)Google Scholar
- T. El-Ghazawi and F. Cantonnet. UPC performance and potential: A NPB experimental study. In Supercomputing 2002 (SC2002), November 2002 Google ScholarDigital Library
- Titanium Project Home Page. http://titanium.cs.berkeley.edu/Google Scholar
- Kaushik Datta et al. Titanium Performance and Potential: An NPB Experimental Study, 18th International Workshop, LCPC 2005, pp. 200--214, Springer, Heidelberg (2006) Google ScholarDigital Library
- Omni Compiler Project. NPB2.3-omni-C, http://www.hpcs.cs.tsukuba.ac.jp/omni-openmp/Google Scholar
- T2K Open Supercomputer. http://www.open-supercomputer.org/Google Scholar
Index Terms
- XcalableMP implementation and performance of NAS Parallel Benchmarks
Recommendations
Implementation and Performance Evaluation of XcalableMP: A Parallel Programming Language for Distributed Memory Systems
ICPPW '10: Proceedings of the 2010 39th International Conference on Parallel Processing WorkshopsAlthough MPI is a de-facto standard for parallel programming on distributed memory systems, writing MPI programs is often a time-consuming and complicated process. XcalableMP is a language extension of C and Fortran for parallel programming on ...
Performance characteristics of openMP constructs, and application benchmarks on a large symmetric multiprocessor
ICS '03: Proceedings of the 17th annual international conference on SupercomputingWith the increasing popularity of small to large-scale symmetric multiprocessor (SMP) systems, there has been a dire need to have sophisticated, and flexible development and runtime environments for efficient and rapid development of parallel ...
Preliminary Implementation of Coarray Fortran Translator Based on Omni XcalableMP
PGAS '15: Proceedings of the 2015 9th International Conference on Partitioned Global Address Space Programming ModelsXcalableMP (XMP) is a PGAS language for distributed memory environments. It employs Coarray Fortran (CAF) features as the local-view programming model. We implemented the main part of CAF in the form of a translator, i.e., a source-to-source compiler, ...
Comments