1. NET0183 Networks and Communications Lectures 29 and 30 Routing and routing protocols... 8/25/20091 NET0183 Networks and Communications by Dr Andy Brooks Lecture powerpoints from the recommended textbook are by Lami Kaya, LKaya@ieee.org. Lecture powerpoints are © 2009 Pearson Education Inc. Their content has sometimes been edited by Andy Brooks. …the propagation of routing information to create and update forwarding tables

2. 8/25/2009 NET0183 Networks and Communications by Dr Andy Brooks 2 The recommended textbook is Computer Networks and Internets by Douglas E. Comer http://www.coursesmart.com/0136066992/?a=1773944 www.pearson-books.com/student (for additional discounts and offers) http://www.coursesmart.com/0136066992/?a=1773944 www.pearson-books.com/student

3. 1 2 3 0111 value in arriving packet’s header routing algorithm local forwarding table header value output link 0100 0101 0111 1001 32213221 Simplified overview of routing and forwarding 8/25/20093 NET0183 Networks and Communications by Dr Andy Brooks figure from Kurose & Ross router

4. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.4 27.2 Static versus Dynamic Routing In static routing, the forwarding table is created before the system starts to forward packets. In static routing, entries are not changed unless they are manually altered by a system manager. In dynamic routing, the software continuously updates the forwarding table to insure that each datagram follows an optimum route. In dynamic routing, the software communicates with other systems to learn optimum routes to each destination. –the software continually checks for network failures Dynamic routing begins exactly like static routing by loading an initial set of routes into a forwarding table when the system boots.

5. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.5 27.3 Static Routing in Hosts and a Default Route Static routing is straightforward and easy to specify. –it does not require extra route propagation software –it does not consume network bandwidth –no CPU cycles are required to propagate routing information However, static routing is relatively inflexible. –it cannot accommodate network failures or changes in topology Where is static routing used? Most hosts use static routing, especially in cases where the host has one network connection and a single router connects the network to the rest of the Internet. –see Figure 27.1

6. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.6 27.3 Static Routing in Hosts and a Default Route Figure 27.1 (a) A typical connection to the Internet, and (b) the static forwarding table used in each host.

7. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.7 27.3 Static Routing in Hosts and a Default Route As Figure 27.1 shows: A static table with two entries suffices for a typical host. –One entry specifies the address of the directly connected network. –The other entry specifies that router R 1 provides a default route for all other destinations. When an application generates a datagram for a computer on the local net (e.g. a local printer), the first entry in the forwarding table directs IP to deliver the datagram directly to its destination. When a datagram is destined for any other destination in the Internet, the second entry in the table directs IP to send the datagram to the router, R 1.

8. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.8 27.4 Dynamic Routing and Routers Most routers on the Internet use dynamic routing but in some exceptional cases static routing can be used. Look at Figure 27.1 again: We can imagine that the figure corresponds to a small organization that is a customer of an ISP. All traffic leaving the customer's site through router R 1 must travel to the ISP. –The forwarding table in router R 1 can be static. –The forwarding table in R 1 can use a default route just as the forwarding table in a host does.

9. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.9 27.4 Dynamic Routing and Routers Static routing and default routes, however, do not suffice for most routers. When two ISPs interconnect, both need to exchange routing information dynamically. See Figure 27.2 Illustration of an architecture that requires dynamic routing.

10. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.10 27.4 Dynamic Routing and Routers In Fig 27.2: –Router R 1 knows about networks 1 and 3. –Router R 1 does not know about network 2 (no direct connection). –Router R 2 knows about networks 2 and 3. –Router R 2 does not know about network 1 (no direct connection). For this trivial example, it may seem that static routing could be made to work, but the static approach does not scale to handle thousands of networks. –Each time an ISP adds a new customer's network, the information must be passed throughout the Internet. –A manual process is far too slow to accommodate new networks, network failures, and congestion in the Internet.

11. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.11 27.4 Dynamic Routing and Routers Each router exchanges information with other routers. The routing software updates the local forwarding table. Routers exchange information periodically. –Andy asks: how often? In Figure 27.2 routers R 1 and R 2 will exchange routing information. –As a result, routing software in R 2 will install a route to network 1 and routing software in R 1 will install a route to network 2. –If router R 2 crashes, the routing software in R 1 will detect that network 2 is no longer reachable and will remove the route from its forwarding table. –Later, when R 2 comes back on line, the routing software in R 1 will determine that network 2 is reachable again and will reinstall the route.

12. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.12 27.5 Routing in the Global Internet - general principles - A route propagation protocol allows one router to exchange routing information with another. However, such a scheme cannot scale to the entire Internet. –If a router in the Internet attempted to exchange routing information with all other routers, the resulting traffic would overwhelm the Internet. To limit routing traffic, the Internet uses a routing hierarchy. –Routers and networks in the Internet are divided into groups. –All routers within a group exchange routing information. –Then, at least one router (possibly more) in each group summarizes the information before passing it to other groups.

13. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.13 27.5 Routing in the Global Internet - general principles - How large is a group? What protocol do routers use within a group? How is routing information represented? What protocol do routers use between groups? The designers of the Internet routing system did not dictate an exact size, nor did they specify an exact data representation or protocol. The designers of the Internet purposefully kept the architecture flexible enough to handle a wide variety of organizations.

14. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.14 27.6 Autonomous System Concept - general principles - The term Autonomous System (AS) is used to specify a group of routers. An AS is a contiguous set of networks and routers all under control of one administrative authority. There is no exact meaning for administrative authority. An AS can correspond to an ISP, an entire corporation, or a university. A large organization with multiple sites may choose to define one AS for each site. Each ISP is usually a single AS, but it is possible for a large ISP to divide itself into multiple ASs. The choice of AS size can be made for economic, technical, or administrative reasons.

15. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.15 27.6 Autonomous System Concept - general principles - It may be less expensive for a multi-national corporation to divide into multiple ASs, each of which has a connection to an ISP in a given country, than to act as a single AS with one connection to the rest of the Internet. A protocol may generate excessive routing traffic when used on many routers, so the protocol imposes a limit on the workable size of an AS. –routing traffic might grow as the square of the number of routers

16. n = 1 n = 2 n = 3 A complete,undirected graph has all possible edges. n = 4 8/25/200916 ALG0183 Algorithms & Data Structures by Dr Andy Brooks An n-vertex, complete, undirected graph has n(n-1)/2 edges. (excluding self-loops)

17. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.17 27.7 The Two Types of Internet Routing Protocols All Internet routing protocols are divided into two major categories: –Interior Gateway Protocols (IGPs) –Exterior Gateway Protocols (EGPs) Routers within an AS use an IGP. Several IGPs are available and each AS is free to choose its own IGP. Usually, an IGP is easy to install and operate. The IGP may limit the size or routing complexity of an AS.

18. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.18 27.7 The Two Types of Internet Routing Protocols 27.7.2 Exterior Gateway Protocols (EGPs) A router in one AS uses an EGP to exchange routing information with a router in another AS. EGPs are more complex to install and operate than IGPs. –but EGPs offer more flexibility and lower traffic overhead To reduce traffic, an EGP summarizes routing information from an AS before passing it to another AS. –Andy asks: how is information summarized? An EGP implements policy constraints that allow a system manager to determine exactly what information is released outside the organization.

19. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.19 27.7 The Two Types of Internet Routing Protocols Figure 27.3 illustrates the two-level routing hierarchy used in the Internet by showing two routers in two ASs. R 1 sends summarized information to R 4 and vice-versa using an EGP.

20. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.20 27.7 The Two Types of Internet Routing Protocols 27.7.4 Optimal Routes, Routing Metrics, and IGPs Routing software should find all possible paths and then choose one that is optimal. Although the Internet usually has multiple paths between any source and destination there is no universal agreement about which path is optimal. Criteria for path optimality include: –latency (delay) for real-time conversations the delay should be less than 150ms –jitter (variability in delay) for real-time conversations the jitter should be less than 30ms –packet loss (dropped packets) for real-time conversations packet loss should be less than 1% –bandwidth (throughput) 1Gbps is better than 1Mbps is better than 1Kbps

21. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.21 27.7 The Two Types of Internet Routing Protocols 27.7.4 Optimal Routes, Routing Metrics, and IGPs Typical Internet routing uses a combination of two metrics: –hop count and administrative cost A hop corresponds to an intermediate network (or router). –The hop count for a destination gives the number of intermediate networks on the path to the destination. Administrative costs are assigned manually. –e.g. paths can be manually assigned to have larger hop counts. “Although most Internet routing protocols are designed to use a hop-count metric, it is possible for a network adminstrator to override the metric to enforce a policy.”

22. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.22 27.7 The Two Types of Internet Routing Protocols 27.7.4 Optimal Routes, Routing Metrics, and IGPs IGPs use routing metrics, but EGPs do not. Each AS is free to choose a routing metric. EGP simply finds a path. An EGP cannot report an optimal path because it cannot make meaningful comparisons: –Suppose one AS reports the number of hops along a path to destination D and another AS reports the throughput along a different path to D. –An EGP that receives the two reports cannot choose which of the two paths has least cost because there is no way to convert from hops to throughput.

23. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.23 27.8 Routes and Data Traffic Data traffic for a given destination flows in exactly the opposite direction of routing traffic. Figure 27.4 (below) illustrates the flow of data in response to routing advertisements. routing advertisements ->

24. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.24 27.9 The Border Gateway Protocol (BGP) Most widely used EGP in the Internet is BGP-4. BGP provides routing information at the AS level. –There is no way for BGP to provide details about the routers within each AS on the path. BGP allows the sender and receiver to enforce policies. –a manager can configure BGP to restrict which routes BGP advertises to outsiders Reliable Transport –BGP uses TCP for all communication

25. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.25 27.9 The Border Gateway Protocol (BGP) BGP classifies each AS as: –a transit system if it agrees to pass traffic through to another AS –or a stub system if it does not Traffic passing through on its way to another AS is classified as transit traffic. BGP allows a corporation to classify itself as a stub. –even if it is multi-homed (i.e. a corporation with multiple external connections can refuse to accept transit traffic) BGP provides the glue that holds Internet routing together at the center of the Internet.

26. Slide 6.26 Halsall, Computing Networking and the Internet, 5 th Edition © Pearson Education Limited 2005 Figure 6.25 Routing over the Internet backbone Halsall: another good networks book. BR – boundary router simplified example

27. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.27 27.10 The Routing Information Protocol (RIP) RIP was one of the first IGPs used on the Internet. RIP is designed to be used among routers within an AS. Hop Count Metric –RIP measures distance in network hops. Unreliable Transport –RIP uses UDP to transfer messages among routers. Broadcast or Multicast Delivery –RIP is intended for use over LAN technologies (e.g. Ethernet) –Version 1 of RIP broadcasts messages –Version 2 allows delivery via multicast The chief advantage of RIP is its simplicity – it requires little configuration.

28. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.28 27.10 The Routing Information Protocol (RIP) RIP uses Distance Vector Routing –Each outgoing message contains an advertisement that lists the networks the sender can reach along with a distance to each. Each entry in a RIP advertisement consists of a pair: (destination network, distance) distance is the number of hops to the destination On receiving an advertisement, RIP software uses the list of destinations to update the local forwarding table. –it calculates the shortest routes to all other nodes and updates its own forwarding table to reflect any changes

29. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.29 27.11 RIP Packet Format The RIP message format helps explain how a distance vector routing protocol operates. Figure 27.5 illustrates a RIP update message and shows how each entry contains the IP address of a destination and a distance to that destination. Each entry also has a next hop address. Each entry contains 20 octets (bytes).

30. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.30 27.11 RIP Packet Format Figure 27.5 The format of a RIP version 2 update message.

31. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.31 27.12 The Open Shortest Path First Protocol (OSPF) The RIP message format illustrates the disadvantages of a distance vector routing protocol: –The size of a message is proportional to the number of networks that can be reached. –Sending RIP messages introduces delay. –Delays mean that route changes propagate slowly. –Processing RIP messages consumes many CPU cycles. RIP works well among a few routers, but does not scale well. To satisfy demand for a routing protocol that can scale to large organizations, the IETF devised an IGP known as the OSPF. –The name is derived from the use of Dijkstra's shortest path first algorithm.

32. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.32 27.13 An Example OSPF Graph Figure 27.6 (a) An example topology, and (b) a corresponding OSPF graph

33. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.33 27.13 An Example OSPF Graph Each node in the graph corresponds to a router. An edge in the graph is a connection between a pair of routers. Each pair of routers connected by a network periodically probe one another and then broadcast a link-state message to other routers. All routers receive the broadcast message. –each uses the message to update its local copy of the graph and to recompute shortest paths if necessary Note that OSPF can use metrics to assign costs to paths.

34. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.34 27.15 Intermediate System - Intermediate System (IS-IS) The IS-IS is an IGP which was originally designed by DEC to be part of DECNET V The two protocols (OSFP and IS-IS) are similar. –Both use the link-state approach employing Dijkstra's algorithm to compute shortest paths. –Both protocols require two adjacent routers to periodically test the link between them and broadcast a status message. When the protocols were initially invented, OSPF's openness made it much more popular than IS-IS. –In fact, IS-IS was almost completely forgotten.

35. © 2009 Pearson Education Inc., Upper Saddle River, NJ. All rights reserved.35 27.15 Intermediate System - Intermediate System (IS-IS) Ironically, in the early 2000's, ten years after the protocols were designed, several things changed to give IS-IS a second chance. DEC had dissolved, and IS-IS was no longer considered valuable proprietary property. A newer version of IS-IS was defined to integrate it with IP and the Internet. The largest ISPs have grown to a size where the extra overhead in OSPF makes IS-IS more attractive. –Over time, OSPF gained many features. As a result, IS-IS has started to make a comeback.

36. 36 http://en.wikipedia.org/wiki/Internet_Protocolhttp://en.wikipedia.org/wiki/Internet_Protocol 29 March 2010 8/25/2009 NET0183 Networks and Communications by Dr Andy Brooks