Modeling and analysis of computer system availability
by A. Goyal, S. S. Lavenberg
The quantitative evaluation of computer-system availability is becoming increasingly important in the design and configuration of commercial computer systems. This paper deals with methods for constructing and solving large Markov-chain models of computer-system availability. A set of powerful high-level modeling constructs is discussed that can be used to represent the failure and repair behavior of the components that comprise a system, including important component interactions, and the repair actions that are taken when components fail. If time-independent failure and repair rates are assumed, then a time-homogeneous continuous-time Markov chain can be constructed automatically from the modeling constructs used to describe the system. Markov chains having tens of thousands of states can be readily constructed in this manner. Therefore, techniques that are particularly suitable for numerically solving such large Markov chains are also discussed, including techniques for computing the sensitivities of availability measures with respect to model parameters. A computer system modeling example is presented to illustrate the use of these modeling and analysis techniques. The modeling constructs, automatic Markov-chain construction, and model-solution methods have been implemented in a program package called the System Availability Estimator (SAVE).