Book Review DOI 10.1007/s001860400382 Wayne D. Grover, (2003) Mesh-Based Survivable Networks. Prentice Hall, Upper Saddle River, ISBN 0-13-494576-X

Survivability is a key aspect for most telecommunication infrastructures. Mesh-based networks are a good basis since they allow different kind of resilience mechanisms. While a tree cannot recover even from a single link failure and a ring offers only one specific backup path, a mesh might be able to react much more flexibly to failures. And failures do occur from time to time (in fact more often than one might think) in all large networks, as some prominent examples stated in the book show. Thus a survivable, yet costefficient network structure that allows a smooth operation even if some failure might have occurred is of great importance for most network operators. The book analyses the backgrounds and shows concepts how to achieve this goal. It is composed of two parts, the so-called ‘‘preparations’’ containing the chapters one to four and the so-called ‘‘studies’’ that constitute the largest part of the book. The ‘‘preparations’’ include a brief introduction into the technological background of current networks. Despite the introductory part, the book is not intended for total beginners in the field of optical networking. This becomes clear right in the first chapter where lots of technical abbreviations are used without any prior explanation. In the following chapters, some technical basics are briefly introduced but the focus quickly turns back towards more advanced topics. This is in no way a criticism of the book, quite the contrary, but something a potential reader should have in mind. A thorough introduction in the field of graph theory, routing and optimization concludes the first part of the book. Beginning with Chapter 5, the main topics of the book start. The first two chapters of this part are devoted to span-restoration/protection and pathrestoration/protection. They introduce the underlying concepts and explain different mathematical models and algorithms for the different types of optimization problems. Chapter 7 deals with oversubscription-based design of Multiprotocol Label Switching (MPLS) and Asynchronous Transfer Mode (ATM) networks. Chapter 8 discusses the effects of dual failures, an aspect often ignored in other publications. Grover shows that it is nevertheless something worth thinking about and that limited protection against dual failures may even be achieved at virtually no extra cost. The impacts of shared risk link groups, a very important aspect for most larger network operators, are also treated in detail in this chapter. Chapter 9 is about ‘‘mesh network design topology and evolution’’. A large part of the book (Chapter 10) is devoted to the concept of p-cycles, a novel ring/mesh hybrid network originally invented by Grover. Consequentially, this chapter gathers the latest research on this topic in an amount that cannot be found anywhere else. It is probably the only comprehensive treatment of this topic so far. The book concludes with a chapter about ring-mesh hybrids and the evolution from existing ring networks towards meshes. Though the book covers a large part of current mesh network planning aspects, it has a clear focus on the topics stated above. Other aspects, like, e.g., routing and wavelength assignment, are only briefly mentioned. Also the word ‘‘mesh’’ in the title is to be taken seriously. Pure rings, trees, stars or any other structures that are not a mesh in the true sense of the word are not

526

Book Review

included. Furthermore the book does not go deep into technical details. To know exactly how, e.g., an optical amplifier works, it is definitely not the one to look in. Also new developments like the Optical Transport Network (OTN) are barely mentioned. Readers who are familiar with the papers published by Grover during the last years will encounter many of them while reading the book. Yet one has not the feeling just to read a compilation of papers, they smoothly fit into the general scope of the book. The book has a large bibliography, but in some places one might wish to have some more references with regard to the latest research or some references that do not belong to Grover’s own publications. A kind of glossary might be helpful even to those experienced in the field of optical networking. It is sometimes difficult to trace back the original description of a term or an abbreviation through the index alone. Besides the book itself, Wayne Grover’s website offers a variety of valuable additional resources that correspond to the chapters of the book like exercises, AMPL files, Errata and so on. Altogether, ‘‘Mesh-Based Survivable Networks’’ by Wayne Grover is a high quality book that delivers a vast amount of up to date information on many topics of current interest in the area of network planning. It is highly recommended as a valuable source for people working in the area of telecommunications network design. Holger Ho¨ller, Hamburg