Application of Relational Interval Arithmetic to Computer Performance Analysis: a Survey |
| |
Authors: | Shikharesh Majumdar |
| |
Affiliation: | (1) Telecommunications Research Institute of Ontario, Department of Systems and Computer Engineering, Carleton University, Ottawa, K1S 5B6, Canada |
| |
Abstract: | This paper presents a survey of the existing work in the area of interval-based performance analysis of computing systems. Intervals in performance analysis is required when uncertainties or variabilities exist in the workload parameters for the performance model of the system. Intervals are also useful for computing upper and lower bounds on system performance. Most conventional analytic models accept a set of single valued parameters and produce a single valued model output. Adaptation of these existing models to handle interval parameters require new techniques that use interval arithmetic. Experiences with relational interval arithmetic provided by a constraint logic programming language in solving a number of performance analysis problems in conventional multiprogrammed computers as well as distributed processing systems are described. |
| |
Keywords: | performance evaluation computer system performance software performance interval-based performance analysis queueing networks performance bounds |
本文献已被 SpringerLink 等数据库收录! |