Link-state routing protocols take a different approach. Link-state routing protocols are more like a road map because they create a topological map of the network and each router uses this map to determine the shortest path to each network.
They use the Dijkstra SPF algorithm. OSPF and IS-IS are both link state routing protocols.
The word link is used to represent a physical interface on a router.
Link State Routing Process
1. Each router learns about its own links, its own directly connected networks.
2. Each router is responsible for meeting its neighbors on directly connected networks.
3. Each router builds a Link-State Packet (LSP) containing the state of each directly connected link.
4. Each router floods the LSP to all neighbors, who then store all LSPs received in adatabase.
5. Each router uses the database to construct a complete map of the topology and computes the best path to each destination network.
Hello packets are used to discover any neighbors on the routers links. A neighbor is any other router that is enabled with the same link-state routing protocol. An adjacency is formed when two routers become neighbors and they continue to use hello packets to maintain their adjacency relationship.
Once adjacencies are built Link State Packets (LSP) can be constructed and sent out to all neighbors. Every router that receives and LSP then forwards this information out on interfaces except the receiving interface; this creates a flooding effect.
LSP are not sent periodically. They only need to be sent:
• During initial startup of the router or routing process.
• Whenever there is a change in the topology.
Once all routers have flooded the network with LSP then the routers can start to use the SPF algorithm to construct SPF trees.
Each router constructs its own SPF tree independently from all other routers.
To ensure proper routing, the link-state databases used to construct those trees must be identical on all routers.
Once the SPF tree is created routes can be added to the routing table.
Link State Advantages
1. Creates a topological map
2. Faster convergence due to the fact that each router does not process the LSP until all have been received and quickly forwards each LSP out all non-receiving interfaces.
3. LSP are only sent when a topology changes occurs and it only sends and LSP that directly relates to it.
4. Hierarchical design is used by the concept of areas. OSPF and IS-IS both use areas to help separate parts of networks. This allows for better route summarization and the isolation of routing issues within an area. Multi-area OSPF and IS-IS are part of the CCNP curriculum.
Exception Notes:
1. OSPF routers flood their own links every 30 minutes, this is called a paranoid update.
2. EIGRP does not send periodic updates.
Link State Requirements
1. More memory because of link-state database and SPF tree.
2. More CPU processing because link-state builds a complete topology.
3. More bandwidth used during initial startup but then much less after convergence.
OSPF
• Designed by IETF OSPF Working Group in 1987.
• Two current versions
o OSPFv2 – for IPv4 networks (RFC 1247 and RFC 2328)
o OSPFv3 – for IPv6 networks (RFC 2740)
• Most work completed by John Moy (OSPF's inventor and primary developer)
IS-IS
• Designed by ISO (ISO 10589)
• First developed at DEC as DECnet Phase V.
• Radia Perlman is chief designer.
• Originally designed for OSI networks
• Now available for IP through Integrated IS-IS and Dual IS-IS
• Known as the routing protocol for ISPs and carriers but becoming more accepted by enterprise networks.
Source
No comments :
Post a Comment