The NERSC Sustained System Performance (SSP) Metric
Abstract
Most plans and reports recently discuss only one of four distinct purposes benchmarks are used. The obvious purpose is selection of a system from among its competitors, something that is the main focus of this paper. This purpose is well discussed in many workshops and reports. The second use of benchmarks is validating the selected system actually works the way expected once it arrives. This purpose may be more important than the first reason. The second purpose is particularly key when systems are specified and selected based on performance projections rather than actual runs on the actual hardware. The third use of benchmarks, seldom mentioned, is to assure the system performs as expected throughout its lifetime1, (e.g. after upgrades, changes, and regular use.) Finally, benchmarks are used to guide system designs, something covered in detail in a companion paper from Berkeley's Institute for Performance Studies (BIPS).
- Authors:
- Publication Date:
- Research Org.:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Org.:
- USDOE Director. Office of Science. Office of AdvancedScientific Computing Research
- OSTI Identifier:
- 861982
- Report Number(s):
- LBNL-58868
R&D Project: KX0110; BnR: KJ0102000; TRN: US200620%%676
- DOE Contract Number:
- DE-AC02-05CH11231
- Resource Type:
- Technical Report
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; BENCHMARKS; METRICS; PERFORMANCE
Citation Formats
Kramer, William, Shalf, John, and Strohmaier, Erich. The NERSC Sustained System Performance (SSP) Metric. United States: N. p., 2005.
Web. doi:10.2172/861982.
Kramer, William, Shalf, John, & Strohmaier, Erich. The NERSC Sustained System Performance (SSP) Metric. United States. https://doi.org/10.2172/861982
Kramer, William, Shalf, John, and Strohmaier, Erich. 2005.
"The NERSC Sustained System Performance (SSP) Metric". United States. https://doi.org/10.2172/861982. https://www.osti.gov/servlets/purl/861982.
@article{osti_861982,
title = {The NERSC Sustained System Performance (SSP) Metric},
author = {Kramer, William and Shalf, John and Strohmaier, Erich},
abstractNote = {Most plans and reports recently discuss only one of four distinct purposes benchmarks are used. The obvious purpose is selection of a system from among its competitors, something that is the main focus of this paper. This purpose is well discussed in many workshops and reports. The second use of benchmarks is validating the selected system actually works the way expected once it arrives. This purpose may be more important than the first reason. The second purpose is particularly key when systems are specified and selected based on performance projections rather than actual runs on the actual hardware. The third use of benchmarks, seldom mentioned, is to assure the system performs as expected throughout its lifetime1, (e.g. after upgrades, changes, and regular use.) Finally, benchmarks are used to guide system designs, something covered in detail in a companion paper from Berkeley's Institute for Performance Studies (BIPS).},
doi = {10.2172/861982},
url = {https://www.osti.gov/biblio/861982},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Sun Sep 18 00:00:00 EDT 2005},
month = {Sun Sep 18 00:00:00 EDT 2005}
}