How not to lie with statistics: the correct way to summarize benchmark results
Communications of the ACM - The MIT Press scientific computation series
Characterizing computer performance with a single number
Communications of the ACM
Performance analysis of high-speed computers
Proceedings of the 1989 ACM/IEEE conference on Supercomputing
Computer architecture: a quantitative approach
Computer architecture: a quantitative approach
Measuring computer performance: a practitioner's guide
Measuring computer performance: a practitioner's guide
Parallel Computer Architecture: A Hardware/Software Approach
Parallel Computer Architecture: A Hardware/Software Approach
More on finding a single number to indicate overall performance of a benchmark suite
ACM SIGARCH Computer Architecture News
War of the benchmark means: time for a truce
ACM SIGARCH Computer Architecture News
Integrated Performance Monitoring of a Cosmology Application on Leading HEC Platforms
ICPP '05 Proceedings of the 2005 International Conference on Parallel Processing
Misleading Performance Reporting in the Supercomputing Field
Scientific Programming
Computer Organization and Design: The Hardware/Software Interface
Computer Organization and Design: The Hardware/Software Interface
| Hi-index | 0.00 |
Sustained performance is the amount of useful work a system can produce in a given amount of time on a regular basis. How much useful work a system can achieve is difficult to assess in a simple, general manner because different communities have their own views of what useful work means and because a large number of system characteristics influence its usefulness. Yet, we, as a community, intuitively, and sometimes explicitly, know when a system is more useful than another. On the other hand, we also know when measures do not accurately portray a system's usefulness. This report will review the important concepts of measuring sustained performance, discuss different approaches for doing the measurements and point out some of the issues that prevent our community from developing common, effective measures.