期刊文献+

HPCC:面向存储访问模型的基准测试—一种可能替代TOP500 HPL的测试方法 被引量:1

HPCC: a Memory Access Model Oriented Benchmark──a Potential Test Method to Replace HPL in TOP500
在线阅读 下载PDF
导出
摘要 高性能计算机系统的性能评价历来是本领域所关注的重要问题.TOP500排名所采用的标准测试HPL(HigPerformanceLinpack)并不能真实的反映系统各方面的性能,尤其是存储访问方面.HPCChallenge基准测试则着重于各种存储访问模型,在HPL的基础之上又整合了多个有代表性的核心测试程序,很有可能在未来取代现在TOP500采用的的HPL测试.本文首先简单介绍HPCChallenge诞生的背景,解释基准测试的基本概念和原理,从存储访问模型的角度对各项测试进行了描述,并根据实际的测试结果进行比较和分析.最后给出结论以及将来的工作. To evaluate the performance of High Performance Computing systems has been an important issue in HPC field. The HPL benchmark, which is used for TOP500 List as a standard to rank the supercomputers all over the world, may not be able to reflect the actual performance for a wide range of applications. The HPCC benchmark, a test suite that stresses on memory access patterns and augments the HPL benchmark with several other representative kernel routines, is potentially to replace HPL for ranking TOP500 List in the near future. This article first introduced the basic concept and discipline of benchmarks, described each kernel test from the viewpoint of the memory access pattern, and then compared and analyzed various test resuits. Finally, a conclusion was given, as well as the future directions for further development of HPCC benchmark.
出处 《小型微型计算机系统》 CSCD 北大核心 2006年第5期950-955,共6页 Journal of Chinese Computer Systems
基金 上海高校网格-E研究院资助
关键词 HPC CHALLENGE 基准测试 高性能计算 局部性 HPC Challenge benchmark high performance computing locality
  • 相关文献

参考文献20

  • 1Top500 supercomputer sites[EB/OL].http://www.top500.org.
  • 2Petitet A, Whaley R C, Dongarra J, et al. HPL-a portable implementation of the high-performance linpack benchmark for distributed-memory computers [EB/OL]. http ://www. netlib.org/benchmark/hpl/.
  • 3HPC challenge benchmark[EB/OL].http://icl.cs. utk. edu/hpcc.
  • 4Cybenko G, Kipp L, Pointer L, et al. Supercomputer performance evaluation and the Perfect benchmark[C]. Proceedings of the 1990 International Conference on Supercomputing,September 1990,254-266.
  • 5James Bunch, Jack Dongarra, Cleve Moler, et al. LINPACK users' gulde[M]. SIAM,Philadelphia:PA,1979.
  • 6NAS parallel benchmark suite [EB/OL]. http://www.nas.nasa. gov/Software/NPB.
  • 7Rudolf Eigenmann, Greg Gaertner, Wesley Jones, et al. SPEC HPC2002:the next high-performance computer benchmark[C].Proceedings of ISHPC 2002, May 2002:7-10.
  • 8Dongarra J J, Hey T. The ParkBeneh benchmark collection[J].Supercomputer, 1995,11 (2-3) : 94-114.
  • 9Rius Juan M, Ferrando Miguel, Jofre Luis. High-frequency RCS of complex radar targets in real-time[J]. IEEE Transactions on Antennas and Propagation, 1993, 41(9):1308-1319.
  • 10Behr M, Briggs P. Parallel implementation of computational fluid dynamics codes on emerging architectures[C]. In: Proceedings of 6th World Multiconference on Systemics, Cybernetics and Informatics, July, 2002,14:105-110.

同被引文献4

引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部