Customer Reviews

OSPF and IS-IS: Choosing an IGP for Large-Scale Networks

5 star:		(7)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

If you consider yourself a student of routing protocols and enjoy coverage of graph theory from the perspective of its application to link-state routing protocols, this text will certainly be a welcome addition to your library. This book not only provides information regarding 'how' link-state routing protocols work, it also provides information regarding 'why' the link-state routing protocols behave as they do, and why the protocol designers made certain choices in the development of these protocols. While it might seem a daunting task especially to the novice reader to learn about two routing protocols side-by-side, it is this treatment which makes this text so worthwhile. Being able to compare these two protocols and identify their similarities and differences simultaneously will ultimately help the network designer pick the right protocol for the job in a given network environment.

This book goes beyond IGP fundamentals by giving practical advice to the network designer which can assist in the planning and implementation of a scalable IGP deployment. For example, in the chapter on Area Design, the author states that "a useful guideline when designing a network is that network control traffic should never exceed 5 percent of the available bandwidth of any link in the network, and in normal circumstances should not exceed 1 percent". The author then presents various formulas which can be used to determine the amount of bandwidth used by the protocol control traffic based on the number and type of LSAs which are expected to be present in a given network. Arguably one of the best chapters in the book is the chapter on Scaling. This chapter has some of the best coverage of the various modifications which router vendors make to their link-state protocol implementations in order to make routers perform calculations more rapidly, enhance flooding of Link-State updates, and other changes designed to make the protocols scale to support very large networks.

I am a stickler for accuracy, especially when it comes to technical textbooks. I pride myself on my ability to spot technical and grammatical errors in texts such as these, however I must say as I read this book I was very impressed that I found very little errors beyond just the simple grammatical and typographical. Jeff Doyle is an experienced writer, and it should come as no surprise that the technical content in this book is extremely well-vetted, accurate, and error-free. Ultimately, if you are a network operator, designer or architect and are interested in broadening your understand of link-state protocols coupled with the ability to more fully understand the technical distinctions between OSPF and IS-IS, this book is without a doubt one of the best options on the market today.

After thoroughly reading the OSPF chapter of Jeff Doyle's Cisco Press title, I scored 100% on the OSPF section of my CCNP Routing exam. This great books gives you the same great information, but presented in a different manner teaching you OSPF/IS-IS on both Cisco and JUNOS side by side. After reading this book, I am confident I could score 100% again.

If you're a hardcore OSPF fan, you will love this book.

OSPF and IS-IS : Choosing an IGP for Large-Scale Networks is an excellent source for understanding the similarities and differences of IS-IS and OSPF all in one book. It is the only book available with such complete coverage comparing these two IGP routing protocols.

There have been many official and unofficial debates over which protocol is better. Finally, the common problem of choosing one IGP routing protocol over the other or the second guessing of an IGP deployment choice can now be put to rest thanks to this definitive source of information on the subject. The book does a fair and deep comparison of the two protocols down to the packet structures and usage scenarios. It answers lingering questions and corrects common misconceptions about how these protocols operate. The author shows no bias towards either protocol without good justification. Jeff presents the information in such a way that the reader can draw his/her own conclusions.

Doyle's book does not assume the reader has strong knowledge of OSPF and IS-IS. It teaches the theory of each protocol in addition to comparing and contrasting in Doyle's easy to follow style. For those who are experts at IS-IS and OSPF with regards to IPv4 and understand the differences, IS-IS and OSPF for IPv6 are covered/compared in the same fashion.

Thanks for a great reference book, Jeff!

In this volume you can tell that Jeff Doyle is getting better and better at conveying complex concepts in an entertaining and informative way. This book is a main stay in my CCIE preparation regarding the link state protocols. I've advised this book to several people, all of which have been very pleased.

I very much enjoy the granular detail of both protocols. Such detail into the functions of each protocol truly allows the engineer to select the right protocol for the job.

Any advanced computer library collection specializing in programmer guides for network designers will want OSPF and IS-IS: Choosing an IGP for Large-Scale Networks: it covers large-scale systems operating in the real world and is organized to help network engineers and architects compare OSPF and IS-IS. From understanding message types and improving scalability to designing large-scale networks for maximum security, OSPF AND IS-IS is a top pick for any in-depth computer library seeking lasting references.

Doyle gives us a very understandable discourse on the OSPF and IS-IS routing methods. These are implemented by Cisco and Juniper routers. Cisco dominates the networking arena, while Juniper is one of the larger secondary players. So understanding OSPF and IS-IS is vital if your duties involve administering networks using those companies' devices.

The link state nature of the OSPF and IS-IS protocols is shown to scale much more easily to large networks, as compared to vector protocols. The latter are slower to converge and are susceptible to looping.

Interestingly, the book starts off with a detailed chapter on the rise of the Internet. It mentions luminaries like Vinton Cerf, Licklider, Kleinrock, Postel and others. And how the ARPANET was the predecessor of the Internet. However, I do take issue with the claim that the Internet began in 1983, when the ARPANET transitioned to TCP/IP. The chapter itself says that "almost all the internetworking technologies we use to this day had their start with the ARPANET." Thus, others who were involved in establishing the ARPANET take the Internet's true beginning to be that of the ARPANET. For example, Kleinrock considers the birth date to be in October 1969, when his group made the first connection between two nodes on the ARPANET, at UCLA and Stanford Research Institute. He and UCLA consider this date to be definitive. (Kleinrock has been at UCLA since the 60s.) Granted, there is an element of hometown boosterism here, but I recently heard him give a seminar with a strong technical description of the 1969 event, and it seemed very plausible.

I should add that even if you consider Doyle's assertion about the Internet's start to be wrong, it does not detract from his first chapter or the rest of the book. The objective events in that chapter are correctly recounted, and the chapter is useful in showing how all this Internet "thing" came about. Read it as good cultural background.

Doyle does an excellent job of starting from ground zero and building up...

Included are points of interest and concerns that a senior level net engineer might think of but someone beginning or intermediate may not quickly consider...

I also like the fact that he compares and contrasts the two protocols side by side in the same section, not doing the 1/2 the book on ospf, then the other half on is-is or chapter specific. Each chapter will talk about a particular subject (say message types) and in the chapter he'll go through both protocols.

Doyle also points out differences in IOS vs JUNOS which is real nice as well.

Chapters are:

I Roots of Link State Protocols: brief history of arpanet (not like IRA 2nd edition either, IRA = Internet Routing Architectures)

II Link State Basics: He does cover DV basics as well to have something to contrast to...

III Message Types

IV Addressing, Neighbor Discovery, Adjacencies

V Flooding

VI Link State Database Synchronization

VII Area Design

VIII Scaling

IX Security/Reliability

X Extensibility

XI MPLS Traffic Engineering

XII IV6

XIII Multi-Topology Routing

Much better than the Cisco Press books and I have all three:

Cisco Press, OReilly, and Addison Wesley...
Although Doyle's TCP/IP Vol II is CP