A TOOLSET FOR PERFORMANCE ENGINEERING AND SOFTWARE-DESIGN OF CLIENT-SERVER SYSTEMS

Citation
G. Franks et al., A TOOLSET FOR PERFORMANCE ENGINEERING AND SOFTWARE-DESIGN OF CLIENT-SERVER SYSTEMS, Performance evaluation, 24(1-2), 1995, pp. 117-136
Citations number
33
Categorie Soggetti
Computer Sciences","Computer Science Hardware & Architecture","Computer Science Theory & Methods
Journal title
ISSN journal
01665316
Volume
24
Issue
1-2
Year of publication
1995
Pages
117 - 136
Database
ISI
SICI code
0166-5316(1995)24:1-2<117:ATFPEA>2.0.ZU;2-3
Abstract
TimeBench/SRVN is a prototype toolset for computer-aided design and pe rformance analysis of software, with an emphasis on distributed client -server systems. The performance behaviour of such systems may defy in tuition because it involves factors in the software design (such as th e partitioning of the functionality and the frequency with which reque sts will be made to each server) and in the configuration of the distr ibuted system (including replication of services, the distribution of data, and the speed of network access). The novelty of the tool consis ts in providing support both for developing design specifications and also for performance analysis. The integrated approach avoids the sema ntic gap between a designer's domain and the performance modeling doma in, and assists the designer to explore factors that impact the perfor mance of a design.The performance models are based on the Stochastic R endezvous Network (SRVN) formalism for client-server systems with sync hronous service requests. The distinctive features of SRVNs are nested services (since servers can also act as clients to other servers) and the existence of two or more phases of service (the first executed wh ile the client is blocked, and the others executed in parallel with th e client). TimeBench/SRVN is intended as a demonstration of the concep t of an integrated designer/performance interface, and as a research e nvironment for fast analytic solvers for the models. Besides a simulat ion solver, it offers three approximate analytic solvers based on rece nt research, a Markovian solver, a technique for finding bounds on the throughput without too many assumptions, and a tool for rapidly explo ring the space of possible parameter values.