iprouting(ppt53)-经营管理(编辑修改稿)

iprouting(ppt53)-经营管理(编辑修改稿)内容摘要：

ction  Uses linkstate packets (LSP), topological database, the SPF algorithm, resulting SPF tree, a routing table of paths and ports to each work  Advantages  Store a single “best” route for each work ID  Scale well to large interworks  Low convergence time  Disadvantages  Complex and difficult to configure  Require additional planning  Examples – OSPF for IP 来自 中国最大的资料库下载 Link State Network Discovery  Routers exchange LSPs with each other.  Each router in parallel with one another constructs a topological database consists of all LSPs  SPF algorithm putes work reachability, determine the shortest path to each other work  With itself as root, the tree expresses path from router to all destinations  Router lists its best paths and ports to all destination works in routing table. It also maintains other databases of topology element and status detail  Routers calculate the shortest path to destination in parallel 来自 中国最大的资料库下载 Link State Topology Changes  Keep track of its neighbors: the name, whether it is up or down and the cost of the link  Constructs an LSP that lists its neighbor router names and link costs. (Includes new neighbors, changes in link cost and link status)  Send out its LSPs to all other routers  When received an LSP, records the LSP in its database and store the most recently generated LSP from each other router  Accumulated LSP data to construct a plete interwork topology map  Rerun the SPF algorithm and pute the route to every work destination 来自 中国最大的资料库下载 Link State Concerns  Processing and memory required for link state routing  Compute the shortest path first with Dijktra’s algorithm requires a processing task proportional to the number of links in the interwork times the number of routers in the work  Bandwidth consumed for initial link state “flood”  Peak demand for bandwidth, temporarily reduces the bandwidth available for carrying user data 来自 中国最大的资料库下载 Link State Problems  Most plex and critical aspect: get all the LSPs necessary  Unsynchronized updates would cause inconsistent path decisions  Synchronizing large works which work topology updates are correct?  Router startup order of start alters the topology learned  Partitioned regions slow updating part separated from fast updating part 来自 中国最大的资料库下载 Link State Solutions  Reduce the need for resources  “Dampen” update frequency  Target linkstate updates to multicast  Use linkstate area hierarchy for topology  Exchange route summaries at area borders  Coordinate linkstate updates  Use time stamps  Update numbering and counters  Manage partitioning using an area hierarchy 来自 中国最大的资料库下载 Routing Information Protocol (RIP) Overview of RIP RIP Timers RIP Configuration RIP Related Configurations 来自 中国最大的资料库下载 Overview of RIP RIP protocol originally RFC 1058  Distance vector routing protocol and hop count is used as metric for path selection  Maximum allowable hop count is 15, meaning that 16 is deemed unreachable  Routing updates broadcast every 30 seconds by default  Work well in small works but inefficient on large works with slow WAN links or on works with a large number of routers installed  RIP ver. 1 – classful routing (all devices in the work must use the same sub mask)  RIP ver. 2 – classless routing, where sub mask information is sent together with the route updates 来自 中国最大的资料库下载 RIP Timers  Route update timer  The interval between periodic routing updates (30 sec by default)  Route invalid timer  The period that the router hasn’t heard any updates about a particular route (90 sec by default)  After that period the route is considered as invalid and the router will send out update about this invalid route to all its neighbors  Route flush timer  The period between a route being invalid and its removal from the routing table (240 sec by default)  Route flush timer Route invalid timer so enough time to tell its neighbors about the invalid route before the routing table is updated 来自 中国最大的资料库下载 RIP Configuration  Router(config)  router rip to start the RIP routing configuration mode  Router(configrouter)  work worknumber to select participating attached works (not specific sub but the whole work with the work number)  passiveinterface interface no. to stop RIP updates from being propagated out from that interface, but it can still receive RIP updates 来自 中国最大的资料库下载 RIP Related Configuration。