A SUBSYSTEM-ORIENTED PERFORMANCE ANALYSIS METHODOLOGY FOR SHARED-BUS MULTIPROCESSORS

Authors
Citation
Cs. Lee et Tm. Parng, A SUBSYSTEM-ORIENTED PERFORMANCE ANALYSIS METHODOLOGY FOR SHARED-BUS MULTIPROCESSORS, IEEE transactions on parallel and distributed systems, 7(7), 1996, pp. 755-767
Citations number
27
Categorie Soggetti
System Science","Engineering, Eletrical & Electronic","Computer Science Theory & Methods
ISSN journal
10459219
Volume
7
Issue
7
Year of publication
1996
Pages
755 - 767
Database
ISI
SICI code
1045-9219(1996)7:7<755:ASPAMF>2.0.ZU;2-Z
Abstract
A methodology, called Subsystem Access Time (SAT) modeling, is propose d for the performance modeling and analysis of shared-bus multiprocess ors. The methodology is subsystem-oriented because it is based on a Su bsystem Access Time Per Instruction (SATPI) concept, in which we treat major components other than processors (e.g., off-chip cache, bus, me mory, I/O) as subsystems and model for each of them the mean access ti me per instruction from each processor. The SAT modeling methodology i s derived from the Customized Mean Value Analysis (CMVA) technique, wh ich is request-oriented in the sense that it models the weighted total mean delay for each type of request processed in the subsystems. The subsystem-oriented view of the proposed methodology facilitates divide -and-conquer modeling and bottleneck analysis, which is rarely address ed previously. These distinguishing features lead to a simple, general , and systematic approach to the analytical modeling and analysis of c omplex multiprocessor systems. To illustrate the key ideas and feature s that are different from CMVA, an example performance model of a part icular shared-bus multiprocessor architecture is presented. The model is used to conduct performance evaluation for throughput prediction. T hereby, the SATPls of the subsystems are directly utilized to identify the bottleneck subsystem and find the requests or subsystem component s that cause the bottleneck. Furthermore, the SATPls of the subsystems are employed to explore the impact of several performance influencing factors, including memory latency, number of processors, data bus wid th, as well as DMA transfer.