Enhanced descriptions from Syndetics:

Quantitative computer performance analysis consists of discovering and ascertaining the efficiency of a computer system; it may be, for example, concerned with the estimation of the performance behaviour of a system under construction, or monitoring that of an existing one. This book presents a body of proven methods on quantitative performance analysis developed over the years, and is based on lecture material given to students of computer science at London and Reading Universities. A large proportion of computer science students feel uneasy about quantitative computer systems analysis because they either find the whole subject obscure and difficult to understand, or they find it difficult to relate the sometimes excessively theoretical treatment to concrete performance situations. Written with this in mind, this book emphasises the practical applications and usable results rather than the general techniques of analysis.

Reviews provided by Syndetics

CHOICE Review

This slim volume provides the system designer with a set of tools for analyzing the impact of design decisions on system efficiency. Leung first supplies the setting for the use of the methods that are described in the remainder of the book; he then follows with a basic introduction to the various forms of "probability statements" that are needed to express many of the inexact situations in computer hardware and software systems. Succeeding chapters cover analytic techniques for system analysis and modeling, numerical techniques for summarizing performance data, and techniques for simulating computer systems. Chapter 9 applies the analytic techniques of the preceding chapters to the areas of database and I/O subsystem performance. The Pollaczek-Khintchine formula from queueing theory is derived in Chapter 4 as a measure of the mean waiting time in a queue. The formula is then applied in succeeding chapters to problems such as the fixed head disk problem in the discussion of I/O subsystems. The bibliography points the reader to expanded explanations of numerous derivations and detailed solutions of simplified problems. This text is a good complement for college courses on computer system simulation. J. D. McGregor Murray State University