High-Performance Client/Server: A Guide to Building and Managing Robust Distributed Systems

Citation

Fellows, G. (1998), "High-Performance Client/Server: A Guide to Building and Managing Robust Distributed Systems", Internet Research, Vol. 8 No. 5. https://doi.org/10.1108/intr.1998.17208eaf.007

High-Performance Client/Server: A Guide to Building and Managing Robust Distributed Systems

High-Performance Client/Server: A Guide to Building and Managing Robust Distributed Systems

Chris Loosley and Frank Douglas1998John Wiley & SonsISBN 0-471-16269-8 (paperback, alk paper)pp. xxxii + 752, incl. index$US34.95www.wiley.com/compbooks

Keywords Client-server computing, Distributed data processing

This book is a collection of original contributions from various authors which Loosley has moulded into a bible of software performance engineering, with the help of his co-author Frank Douglas.

There are 22 chapters arranged in six parts:

1. 1.

   Foundations, which give an introduction to client-server computing, especially the enterprise, and performance fundamentals.

2. 2.

   Process, which discusses software performance engineering (SPE) which is really the focus of this book.

3. 3.

   Principles, which describes the six key principles of SPE: "The workload principle", "The efficiency principle", "The locality principle", "The sharing principle", "The sharing principle: shared databases", and "The trade-off principle".

4. 4.

   Applications, showing how the ideas and principles are applied to the design of distributed systems for the enterprise.

5. 5.

   Technologies, which describes performance tools, database management systems, data replication, data warehousing and transaction managers and monitors.

6. 6.

   Resources, which has a performance-oriented glossary and an index of guidelines by subject.

One of the key attributes of this book is a series of guidelines. These offer a recommendation, or conclusion, for most of the design discussions in the book. For example, the guideline at the end of the section on efficiency principles, from Greg Faust of Microsoft, offers a well-held view: first design software that works, then optimize your design. A preceding discussion reiterates Connie Smith's remark that "You can always tune it later, but the results are never as good".

The lack of bibliography or further reading sections is offset by the frequent references and recommendations for further reading in the body of the text. I would highly recommend this book to managers and developers of enterprise-level client-server systems. It is a reference work for that group compared with Linthicum's "Client/server and intranet development" tools-based approach or Ofali et al.'s. "Client/server survival guide". These books overlap slightly. Linthicum contributed the first draft to the chapter on performance tools and in several places Loosley suggests Ofali's work is a comprehensive guide to client-server computing especially middleware.

Geoff Fellows School of Information StudiesCharles Sturt University, Australia