Computer system models

Abstract

A system model is a tool used to predict system performance under changing conditions. There are two widely used modeling techniques: one based on discrete event simulation and one based on queuing theory models. Because queueing theory models are so much cheaper to implement and use, as compared to simulation models, there is growing interest in them. Users are developing and using queuing theory models to project system performance, project capacity, analyze bottlenecks and configure systems.This talk uses an operational analysis approach to develop system models. This approach, as presented in Denning and Buzen [1], provides an intuitive basis for analyzing system performance and constructing system models. Very simple calculations lead to estimates of bounds on performance - maximum job throughput rates and minimum message response times. The emphasis is on gaining an understanding of system models which reenforces intuition, not on mathematical formulae.Several examples are included. References to other works and publications are provided. Application areas and limitations of modeling techniques are discussed.