Architecture independent performance characterization andbenchmarking for scientific applications
A simple, tunable, synthetic benchmark with a performance directly related to applications would be of great benefit to the scientific computing community. In this paper, we present a novel approach to develop such a benchmark. The initial focus of this project is on data access performance of scientific applications. First a hardware independent characterization of code performance in terms of address streams is developed. The parameters chosen to characterize a single address stream are related to regularity, size, spatial, and temporal locality. These parameters are then used to implement a synthetic benchmark program that mimics the performance of a corresponding code. To test the validity of our approach we performed experiments using five test kernels on six different platforms. The performance of most of our test kernels can be approximated by a single synthetic address stream. However in some cases overlapping two address streams is necessary to achieve a good approximation.
- Research Organization:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
- Sponsoring Organization:
- USDOE Director. Office of Science. Office of Computationaland Technology Research
- DOE Contract Number:
- DE-AC02-05CH11231
- OSTI ID:
- 860743
- Report Number(s):
- LBNL-56306; R&D Project: K11114; BnR: KJ0101030; TRN: US200524%%306
- Resource Relation:
- Conference: MASCOTS 2004, Volendam, The Netherlands, October5-7, 2004
- Country of Publication:
- United States
- Language:
- English
Similar Records
Resilient and Robust High Performance Computing Platforms for Scientific Computing Integrity
Science Driven Supercomputing Architectures: AnalyzingArchitectural Bottlenecks with Applications and Benchmark Probes