TCOM 509 – Internet Protocols (TCP/IP) Lecture 06_a Routing Protocols: RIP, OSPF, BGP

1. TCOM 509 – Internet Protocols (TCP/IP)Lecture 06_aRouting Protocols:RIP, OSPF, BGP Instructor: Dr. Li-Chuan ChenDate: 10/06/2003 Based in part upon slides of Prof. J. Kurose (U Mass)

2. Routing in the Internet • The routing tables contains the optimum path for packets • Routing Protocols allow routers to exchange routing info with other to maintain their routing tables. • Autonomous Systems (AS): A collection of hosts and routers that are administered by a single authority. • The Global Internet consists of AS interconnected with each other • Two-level routing: • Interior routing (Intra-AS): routing inside an AS • RIP, OSPF • Exterior routing (Inter-AS): routing between two ASs • BGP

3. Internet AS Hierarchy Intra-AS border (exterior gateway) routers (BGP) Inter-ASinterior (gateway) routers (RIP, OSPF)

4. Intra-AS Routing • Also known as Interior Gateway Protocols (IGP) • Most common Intra-AS routing protocols: • RIP: Routing Information Protocol • OSPF: Open Shortest Path First • IGRP: Interior Gateway Routing Protocol (Cisco proprietary)

5. RIP ( Routing Information Protocol) • Distance vector routing (Bellman-Ford Algorithm) • Distance metric: # of hops • max = 15 hops • 16 is infinity (unreachable) • RIP Message Types • Request • Specific entry • All entries • Response • Answer to request • Periodic sent every 30 seconds for all entries in the table (also called advertisement)

6. RIP Timers • Periodic timer: advertisement (update) message sent every 30 sec • Expiration timer: governs the validity of a route, default 180 sec • If not heard within 180 sec, mark hop count to 16 (unreachable) and the route as invalid. • Reset when a new update message is received. • Garbage collection timer: advertises the failure of a route, default 120 sec • Start this timer when the Expiration timer expired. • Remove the entry when this timer expired.

7. RIP Instability Cons • Slow convergence: it takes an average of 15 sec for an update to reach the next router. For n routers, it may take 15 x n sec to propagate an update. (solution: limit the hop count) • Instability: A packet could be in a loop between adjacent routers. Net1 --- R1 --- Net2 --- R2 --- Net3X 1 2 (hop count)X 16 2 X 3 4 … X 16 16 • Solutions to improve RIP Stability: • Triggered update • Split horizon: will not advertise network info out the same interface from which it learned about the network.

8. RIP ( Routing Information Protocol) Version 1 • use broadcast to send msgs to every neighbor. Version 2 • Route Tag: allows RIP to receive info from exterior routing protocol • Subnet Mask: support classless addressing and CIDR • Authentication: prevent unauthorized advertisement • Multicast: sent only to RIP routers. Encapsulation: encapsulated in UDP (port 520).

9. OSPF (Open Shortest Path First) • “open”: publicly available • Uses Link State (LS) algorithm • Each router sends the state of its router in the area. A packets is sent only if there is a change in the area. • Topology map at each node • Route computation using Dijkstra’s algorithm • Encapsulate OSFP message in IP (rather than TCP or UDP)

10. OSPF “advanced” features (not in RIP v1) • Security: all OSPF messages authenticated (to prevent malicious intrusion) • Multiple same-cost paths allowed (only one path in RIP) • For each link, multiple cost metrics for different TOS(delay, throughput, etc. e.g., satellite link cost set “low” for best effort; high for real time) • Integrated unicast and multicast support: • Multicast OSPF (MOSPF) uses same topology data base as OSPF • Scales: hierarchical OSPF in large domains.

11. Hierarchical OSPF

12. Hierarchical OSPF • Two-level hierarchy: local area, backbone. • Link-state advertisements only in area • each nodes has detailed area topology; only know direction (shortest path) to nets in other areas. • Area border routers:“summarize” distances to nets in own area, advertise to other Area Border routers. • Backbone routers: run OSPF routing limited to backbone. • Boundary routers: connect to other AS’s.

13. Inter-AS routing in the Internet: BGP

14. Internet inter-AS routing: BGP • BGP (Border Gateway Protocol):the de facto standard • Path Vector protocol: • similar to Distance Vector protocol • each Border Gateway broadcast to neighbors (peers) entire path (i.e., sequence of AS’s) to destination • Loop Prevention: If its AS is in the path, ignore message (loop occurred). No instability problem like RIP. • BGP routes to networks (ASs), not individual hosts • Metric: based on policy

15. BGP messages • BGP messages exchanged using TCP (port 179). No error control and flow control needed. • BGP supports classless addressing and CIDR.

16. Why different Intra- and Inter-AS routing ? Policy: • Inter-AS: admin wants control over how its traffic routed, who routes through its net. • Intra-AS: single admin, so no policy decisions needed Scale: • hierarchical routing saves table size, reduced update traffic Performance: • Intra-AS: can focus on performance • Inter-AS: policy may dominate over performance