Network routing protocols have emerged as key enabling technologies in a computing world now dominated by connectivity. From a very high-level perspective, these routing protocols can be split into interior gateway protocols (IGPs) and exterior gateway protocols (EGPs). In general, the routing techniques used by IGPs are based on either distance-vector or link-state algorithms.
The Open Shortest Path First (OSPF) routing protocol has evolved into the link-state protocol of choice for many IP networks. This has come about for a variety of converging reasons. Most importantly, OSPF has proved to be both reliable and scalable. In addition, its underlying protocol assumptions encourage a structured network design approach, while these same characteristics promote rapid route convergence during operation. The basic features and capabilities of OSPF are described together as a set of specifications under the Requests for Comment (RFCs) regulated by the OSPF Working Group of the Internet Engineering Task Force (IETF).
From OSPF’s earliest days, Cisco has been closely involved with the evolution of related IETF standards. Throughout this process, Cisco’s development engineering staff has worked carefully to ensure that the implementation of OSPF in Cisco routers is both robust and comprehensive. However, as with any complex network topology, uncontrolled growth without careful network design can lead to performance and convergence problems--even with OSPF. At its core, one of the key objectives of Tom Thomas’ book, OSPF Network Design Solutions, is to help network engineers and architects avoid the pitfalls of unstructured network deployment.
This book aims to provide specific Cisco solutions for network engineers deploying OSPF in large-scale IP networks. In doing so, we hope that it contributes to your information toolkit in a substantive way and facilitates the creation of robust and reliable network infrastructures. While the emphasis here is on OSPF and Cisco’s implementation, we also hope that the ideas presented will help anyone deploying large networks using link-state routing protocols--regardless of the specific underlying protocols or equipment.
