Designing Large-Scale LANs
Large networks are complex systems, with many interdependent devices held together only by the strength of the network designer’s specifications. It takes a lot of knowledge–and a lot of time–to make sure that the capacity of links and nodes is neither wasted nor overloaded.
Designing Large-Scale LANs explains how the pieces of a large local area network (LAN) interoperate. A LAN, in author Kevin Dooley’s working definition here, is a building-size data network, or perhaps a campus-size system with a backbone running Gigabit Ethernet, ATM, or FDDI. Dooley does a good job of teaching his readers about, for example, the processing overhead involved in address translation or protocol tunneling, and why techniques like those should be used sparingly in a LAN that needs to be fast and efficient.
That said, Dooley spends a lot of time on network fundamentals. He explains subnet masking in his own rather appealing way, and takes readers on a tour of the Open Systems Interconnect (OSI) seven-layer model. He explains the function of split horizon and poison reverse in Routing Information Protocol (RIP), as well as the improvements more modern routing protocols make on RIP. In short, there’s an awful lot of background information here. Designers of big networks need to understand this stuff, but you might expect to see comparisons of competing vendors’ equipment here–there’s not enough of that–and less broad, conceptual information. On the other hand, you could do far worse than to study this book in preparation for a networking certification exam. –David Wall
