Abstract
Standard benchmark suites are a popular way to measure and compare computers performance. The Honeywell Technology Center has developed two benchmarking suites for parallel computers, the C3I Parallel Benchmark Suite (C3IPBS) and more recently, the Real Time Parallel Benchmark Suite (RTPBS). The C3IPB suite was similar to other benchmarks but focused on a new domain. The development of the real-time benchmark suite presented a number of new challenges. In this paper we share some of the lessons we have learned about the art of making parallel benchmarks.
This work was sponsored by USAF Rome Laboratory, contract number F30602-94-C-0084.
This work was sponsored by DARPA via USAF Rome Laboratory contract number F30602-94-C-0084.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
“Low-Level Benchmark Specifications”, Technical Information Report (C006), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Low-Level Benchmark Results”, Technical Information Report (C007), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Kernel Level Benchmark Specifications”, Technical Information Report (C008), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Kernel-Level Benchmark Results”, Technical Information Report (C009), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Application Level Benchmark Specifications”, Technical Information Report (C0010), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998.
“Application Level Benchmark Results”, Technical Information Report (C0011), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Benchmarking Methodology”, Technical Information Report (C005), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Benchmark Requirements and Selection”, Technical Information Report (C004), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“Benchmark Distribution Plan Report”, Technical Information Report (C012), Rome Laboratory, Contract Number F30602-94-C-0084, August 1998
“A Real-Time Parallel Benchmark Suite”, B. Van Voorst, L. Pires, R. Jha, Proceedings of SIAM’s Parallel Processing for Scientific Computing, SIAM, March 1997.
“Real-Time Embedded High Performance Computing: Application Benchmarks”, C. Brown, M. Flanzbaum, R. Games, J. Ramsdell, MITRE Technical Report, MTR 94B0000145, October 1994.
“C3I Parallel Benchmark Suite Final Report”, Technical Information Report (A012), Rome Laboratory, Contract Number F30602-94-C-0084, May 1998
“The C3I Parallel Benchmark Suite-Introduction and Preliminary Results”, R. Metzger, B. Van Voorst, L. Pires, R. Jha, W. Au, M. Amin, D. Castanon, V. Kumar, Proceedings from Supercomputing 96, 1996.
“Benchmarking Methodology for Real-Time Embedded Scalable High Performance Computing”, R. Games, MITRE Technical Report, MTR96B0000010, March 1996.
“NAS Parallel Benchmark (Version 1.0) Results 11–96”, Subhash Saini, David Bailey, NASA Ames Technical Report NAS-96-18, November 1996.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1999 Springer-Verlag
About this paper
Cite this paper
Van Voorst, B., Jha, R., Ponnuswamy, S., Pires, L., Nanavati, C., Castanon, D. (1999). Reflections on the creation of a real-time parallel benchmark suite. In: Rolim, J., et al. Parallel and Distributed Processing. IPPS 1999. Lecture Notes in Computer Science, vol 1586. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0098014
Download citation
DOI: https://doi.org/10.1007/BFb0098014
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65831-3
Online ISBN: 978-3-540-48932-0
eBook Packages: Springer Book Archive