undergraduate origins shows up throughout the book. The mathematics is presented very gently. D a v i d F i n k e l For example, several complex formulas are writB o o k R e v i e w E d i t o r ten out twice, once without summation notation (i.e., al + a2 + ... + an) and then again with summation notation ( ~ ai). There are numerous Starting with this issue, Performance Evalua- worked out examples, and a wide range of exertion Review will have a regular column devoted cises, from simple ones that just use the formulas to reviews of performance evaluation books. I in the text to more challenging exercises. had intended to inaugurate the column with The book starts with some introductory chapbrief reviews of all the performance books that ters outlining the area of performance evaluhad appeared recently. However, because of the ation, briefly reviewing important topics from large number of books, including the many sug- computer system architecture and showing how gested to me via the SIGMETRICS Electronic these topics impact performance. The next Bulletin Board, I have divided the books into three chapters discuss queueing models of comtwo categories: general performance evaluation puter systems, using the Operational Analybooks, appropriate for a first course in perfor- sis approach. The authors discuss stand-alone mance evaluation, and more specialized or ad- queues and open and closed networks of queues. vanced books. The first group of books are re- There are separate chapters for models with mulviewed briefly in this issue. The more advanced tiple workload classes and with load-dependent books will be reviewed in the next issue. service centers. A compact chapter on simulaAll of these books, and others as they appear, tion covers the generation of random numbers, will be the subjects of in-depth reviews in later the structure of a simulation program, and the issues of Performance Evaluation Review. If you analysis of simulation output. The chapter ends would be willing to be a reviewer, please let me with a complete program in Pascal to simulate a know. Please send me enough information about closed terminal system. your background and experience so that I can The next two chapters cover workload modelassign books to reviewers appropriately. Also, ing and methods of system improvement, such if you know of relevant books that haven't been as load balancing and system tuning. The fireviewed here, or if you have suggestions about nal three chapters discuss an important topic not how this column could be made more useful to usually addressed in book at this level, capacity you, please let me know. management, even discussing management aspects of of the capacity manager's job. David Finkel Department of Computer Science T h e A r t o f C o m p u t e r S y s t e m s P e r f o r Worcester Polytechnic Institute m a n c e A n a l y s i s , R. JMn, Wiley-Interscience, Worcester, MA 01609 1991. $49.95 dfinkel@cs.wpi.edu I reviewed this book from the author's (508) 831-5416 manuscript. The book is now being typeset, and the author tells me that it is due to appear in February, 1991. General Performance Text Books The author's introduction describes this book C o m p u t e r S y s t e m s P e r f o r m a n c e M a n a g e - as written for "computer professionals who dem e n t a n d C a p a c i t y P l a n n i n g , J. Cady and B. sign, procure, or use computer systems," and Howarth, Prentice-Hail, 1990. 192 pp. $43.20. grew out of the author's experience teaching a This book grew out of the authors' experi- graduate course at M.I.T. It is extremely comence teaching a course in performance evaluation prehensive - I am hard pressed to think of a to final year undergraduates. This heritage of performance evaluation technique that isn't dis- Book Reviews Performance Evaluation Review Vol. 18 #3, November 1990

/lp/association-for-computing-machinery/book-review-the-art-of-computer-systems-performance-analysis-by-r-jain-PdH1absfy3