The quandry of benchmarking broadcasts
A message passing library's implementation of broadcast communication can significantly affect the performance of applications built with that library. In order to choose between similar implementations or to evaluate available libraries, accurate measurements of broadcast performance are required. As we demonstrate, existing methods for measuring broadcast performance are either inaccurate or inadequate. Fortunately, we have designed an accurate method for measuring broadcast performance. Measuring broadcast performance is not simple. Simply sending one broadcast after another allows them to proceed through the network concurrently, thus resulting in accurate per broadcast timings. Existing methods either fail to eliminate this pipelining effect or eliminate it by introducing overheads that are as difficult to measure as the performance of the broadcast itself. Our method introduces a measurable overhead to eliminate the pipelining effect.
- Research Organization:
- Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
- Sponsoring Organization:
- USDOE Office of Defense Programs (DP) (US)
- DOE Contract Number:
- W-7405-ENG-48
- OSTI ID:
- 14472
- Report Number(s):
- UCRL-JC-133177; DP0101031; DP0101031; TRN: AH200136%%692
- Resource Relation:
- Conference: The Eighth International Symposium on High Performance Distributed Computing, Redondo Beach, CA (US), 08/03/1999--08/06/1999; Other Information: PBD: 5 Feb 1999
- Country of Publication:
- United States
- Language:
- English
Similar Records
Line-plane broadcasting in a data communications network of a parallel computer
Line-plane broadcasting in a data communications network of a parallel computer