期刊文献+

面向评估的高产出率计算指标量化分析

Quantified Analysis of High Productivity Computing Metrics Oriented to Evaluation
下载PDF
导出
摘要 为了解决当前传统的高性能计算片面强调性能而忽视系统各方面的平衡性所导致的问题,研究人员提出了"高产出率计算"的概念,并为此展开了高产出率计算评估体系的研究。本文从当前高产出率计算评估体系研究现状出发,从高产出率计算对传统"性价比"评估概念的扩展入手,提出了以"指标量化"和"统一量纲"为中心的评估高产出率计算效果的方案,并针对计算能力、鲁棒性、能耗和易用性四个要素给出了一种量化的评估办法。 In traditional high performance computing area, the system's performance gets the greatest focus. The balance among all aspects of the whole system is neglected. The idea for building computing system causes problems. An approach for evaluation of high productivity computing (alias HPC) based on quantified metrics and unified dimensions is provided, starting with the status quo of the research on evaluation schema of HPC, and the extension to the evalua- tion concept of the traditional ' performance-cost ratio' by HPC. A quantified evaluation method is proposed on 4 main factors of HPC: computing abilities, robustness, energy consumption and the ‘easy-to-use’ abilities.
出处 《计算机科学》 CSCD 北大核心 2007年第11期290-292,共3页 Computer Science
基金 863软件专项-服务器操作系统(项目号2002AA1Z2101)
关键词 高产出率计算 评估指标 量化 量纲 Hing produetivity computing, Evaluation metrics, Quantification, Dimension
  • 相关文献

参考文献6

  • 1孟丹,张志宏,陈明宇.高生产率计算系统[J].计算机研究与发展,2005,42(4):563-569. 被引量:6
  • 2Kepner J.HPC Productivity:An Overarching View.International Journal of High Performance Computing Applications,2004,18(4)
  • 3Sterling T.Productivity Metrics and Models for High Performance Computing.International Journal of High Performance Computing Applications,2004,18(4):433-440
  • 4Kennedy K,Koelbel C,Schreiber R.Defining and Measuring the Productivity of Programming Languages.International Journal of High Performance Computing Applications,2004,18(4):441-448
  • 5Faulk S,Gustafson J,Johnson P,et al.Measuring HPC Productivity.International Journal of High Performance Computing Applications,2004,18(4):459-473
  • 6Gustafson J.Purpose-based Benchmarks.International Journal of High Performance Computing Applications,2004,18(4):475-487

二级参考文献11

  • 1DARPA. High Productivity Computing Systems (HPCS)Program. http: // www. darpa. mil/ipto/Programs/hpcs/index.htm, 2002-04.
  • 2Mootaz Elnozahy. PERCS: IBM effort in HPCS. http: // www.ncsc. org/casc/meetings/vision-public. pdf, 2003.
  • 3Cray Inc. The cascade project. http://www. cray. com/cascade/,2002.
  • 4Per Nyberg. The Cray rainier system: Integrated scalar/vector computing. http://www. ecmwf. int/newsevents/meetings/workshops/2004/high_performance-computing-11th/pdf/Per_Nyberg. pdf, 2004.
  • 5John L. Gustafson. Sun' s HPCS approach: Hero. http://www.ncsc. org/casc/meetings/CASC2. pdf, 2003-08.
  • 6HPCC. HPC challenge benchmark. http://icl. cs. utk. edu/hpcc/,2004.
  • 7J. Gustafson. Purpose-based benchmarks. http: // www.highproductivity. org/IJHPCA/8f-Gustafson-Productivity. pdf,2004.
  • 8Thomas Sterling. Productivity metrics and models for high performance computing. http: // www. highproductivity. org/IJHPCA/4f-Sterling-ProductivityPaperv3. pdf, 2004.
  • 9John D. McCalpin. Issues in the architecture of large scale systems. http://www. spscicomp. org/ScicomP10/Presentations/McCalpin. pdf, 2004-09.
  • 10Kemal Ebcioglu, Vijay Saraswat, Vivek Sarkar. X10: New opportunities for compiler-driven performance via a new programming model. http: // www. cs. ualberta. ca/~ amaral/cascon/CDP04/slides/sarkar. pdf, 2004-10.

共引文献5

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