Introduction to Internetworking: Connecting Networks and Addressing

1. Chapter 4 • “There is more than one network” • Point-to-point links • Shared media • Switches • Goal – connect these things together • Heterogeneity • Scale • Routing(connection of nodes) • Addressing • Introduce the Internet Protocol(IP) Internetworking

2. Network 1 (Ethernet) H7 R3 H8 H1 H8 H2 H1 H3 TCP TCP Network 4 R1 R2 R3 (point-to-point) Network 2 (Ethernet) R1 IP IP IP IP IP R2 FDDI PPP ETH ETH ETH FDDI PPP ETH H4 Network 3 (FDDI) H5 H6 Internet, internet or internetwork • Concatenation of Networks • Protocol Stack Routers/Gateways Internetworking

3. Service Model • the services that are to be provided from host-to-host • IP provides Datagram data delivery service • Connectionless (datagram-based) / full dest • Best-effort delivery (unreliable service) • packets are lost • packets are delivered out of order • duplicate copies of a packet are delivered • packets can be delayed for a long time Internetworking

4. 0 4 8 16 19 31 TOS Length V ersion HLen Ident Flags Offset TTL Protocol Checksum SourceAddr DestinationAddr Pad Options (variable) (variable) Data Datagram Version-IP e.g. IPv4, IPv6 Hlen-Header length TOS-Type Of Service(ch6) TTL-Time To Live Protocol-UDP/TCP • Datagram format Assigned at 32-bit Boundary Internetworking

5. Fragmentation and Reassembly • Each network has some Maximum Transmission Unit(MTU) – largest IP datagram that it can carry in a frame. • Strategy • fragment when necessary (MTU < Datagram) • try to avoid fragmentation at source host • re-fragmentation is possible when travel to a different network • fragments are self-contained datagrams • delay reassembly until destination host • do not recover from lost fragments Internetworking

6. Network 1 (Ethernet) H7 R3 H8 H2 H1 H3 Network 4 (point-to-point) Network 2 (Ethernet) R1 R2 H4 Network 3 (FDDI) H5 H6 Global Addresses • All hosts attached to the same network have the same network field • The host field unique • Routers have two interfaces Internetworking

7. Datagram Forwarding • How do IP routers forward datagrams? • Strategy (main points) • every datagram contains destination’s address • if directly connected to destination network, then forward to host(i.e. same network number) • if not directly connected to destination network, then forward to some router(next hop router-host selected) • Hosts = 1 interface, routers >= 2 interfaces • forwarding table maps network number into next hop • each host has a default router • each router maintains a forwarding table Internetworking

8. Network 1 (Ethernet) H7 R3 H8 H2 H1 H3 Network 4 (point-to-point) Network 2 (Ethernet) R1 R2 H4 Network 3 (FDDI) H5 H6 Datagram Forwarding • Example (R2) Network Number Next Hop 1 R3 2 R1 3 interface 1 4 interface 0 0 1 Routers/Gateways Internetworking

9. Bridges, Switches, and Routers • Confusion? • Bridges • Link-level to forward frames from one link to another to create extended LANs • Switches • Network-level to forward packets from one link to another to create packet-switched networks • Routers • Internet-level to forward datagrams from network to another • Bridges/Switches (above the physical below the internet) Internetworking

10. IP Mapping • IP datagrams contain IP addresses • Physical interface only understands the addressing specific to its link-level(i.e. Ethernet, Token-Ring, etc) • Problem – how to put the two together? • Solution – have each host maintain a table of address pairs. Internetworking

11. ARP - Address Translation • Map IP addresses into physical addresses • destination host • next hop router • ARP cache or ARP table • ARP (Address Resolution Protocol) • table of IP to physical address bindings dynamically created • broadcast request if IP address not in table • target machine responds with its link-level and physical address • table entries are discarded if not refreshed Internetworking

12. DHCP • Why Dynamic Host Control Protocol (DHCP) ? • Manual configuration is problematic • incorrect network# and IP address • DHCP Server • Contains list of IP addresses in a range(same network number) • IP addresses are leased • Use DHCPDISCOVER(255.255.255.255) broadcast • Server responds by filling in corresponding IP field of packet sent back to client Internetworking

13. ICMP - Error Reporting • Internet Control Message Protocol(ICMP) reports • Link failure • Reassembly failure • TTL = 0 • CRC failed • IP & ICMP always configured together Internetworking

14. Routing • For VC’s (one-off setup) • For datagrams, every packet must be routed • Regardless, switches and routers must • Look at destination port • Determine the output port best suited transmission Internetworking

15. Routing & Forwarding • Routing tables are constructed as precursors to forwarding table construction • Routing algorithms • Map Ntwk #s to NHs • Optimized for changes in topology • Forwarding tables • Ntwk #s, Output interfaces, MAC address of NH • Optimized to look up Ntwk #s Internetworking

16. Routing & Forwarding • Routing Table NetworkNH (NextHop) 10 171.29.222.38 • Forwarding table NetworkIntMAC Address 10 if0 8:0:2 B:E 4:B:1:2 Internetworking

17. Routing Domain • What is a routing domain? it is a matter of graph theory hosts, switches, routers, or networks network links w/associated costs Internetworking

18. Routing Protocol • How to store the shortest paths? calculate and store in non-volatile memory? • Nope! • Node/link failures not handled • Node/link additions not handled • Edge costs do not change • Run routing protocols among nodes instead • Distance vector and link state Internetworking

19. Distance Vector(Bellman-Ford) • Each node maintains a set of triples • (Destination, Cost, NextHop) • Exchange updates directly connected neighbors • periodically (on the order of several seconds) • whenever table changes (called triggered update) • Each update is a list of pairs: • (Destination, Cost) • Update local table if receive a “better” route • smaller cost • came from next-hop • Refresh existing routes; delete if they time out Internetworking

20. Distance Vector Destination Cost NextHop B 1 B C 1 C D E 1 E F 1 F G • Destination Cost NextHop • B 1 B • C 1 C • D 2 C • E 1 E • F 1 F • G 2 F • With no topological changes • Only a few exchanges for completion • Realization of routing tables at nodes is called convergence • The beauty - no one node is responsible for being the central authority Internetworking

21. Routing Updates • Two circumstances under which a node sends an update • Periodic updates (I’m alive!) • Triggered updates (recipient) Internetworking

22. Routing Failure • Example 1: IF LINK from F to G WENT DOWN • F detects that link to G has failed • F sets distance to G to infinity and sends update t o A • A sets distance to G to infinity since it uses F to reach G • A receives periodic update from C with 2-hop path to G • A sets distance to G to 3 and sends update to F • F decides it can reach G in 4 hops via A • Example 2: PROBLEM of count to infinity • link from A to E fails • A advertises distance of infinity to E • B and C advertise a distance of 2 to E • B decides it can reach E in 3 hops; advertises this to A • A decides it can read E in 4 hops; advertises this to C • C decides that it can reach E in 5 hops… Internetworking

23. Count to Infinity • Solutions • Set max at 16 • What if the network grows • Stabilization (two nodes) • Split Horizon (don’t send routes learned from neighbors) • Split Horizon with Poison Reverse • Problem – speed of convergence • Link-state routing Internetworking

24. Routing Protocols • Interior Gateway Protocols(IGP) • Distance Vector • Updates based upon directly connected neighbors • Destination, cost, and NH (NextHop) • Count-to-Infinity • RIP • Link State • Updates communicated through network Internetworking

25. Routing Information Protocol(RIP) • Probably the most used routing protocol today • Example distance vector protocol • RIP send advertisements every 30 seconds • Routers also send updates when it receives an update from another router • All link costs = 1 • Minimum hop route • Valid distance = 1 16() Internetworking

26. Link State Routing • Strategy • send to all nodes (not just neighbors) information about directly connected links (but not entire routing table) • Reliable flooding of Link State Packet (LSP) • Each node then has complete topology of network • Apply any shortest path algorithm (e.g. Dijkstra) to find the shortest route • Link State Packet (LSP) • id of the node that created the LSP • cost of link to each directly connected neighbor • sequence number (SEQNO) • time-to-live (TTL) for this packet Internetworking

27. OSPF • Open Shortest Path First (OSPF) • Most commonly used type of LSP Internetworking

28. Distance Vector vs. Link State Internetworking

29. Tree Structured Internet(c. 1990) NSFNET backbone Stanford ISU BARRNET MidNet … regional regional W estnet regional Berkeley P ARC UNL KU UNM NCAR UA Autonomous Systems Internetworking

30. The Global Internet • We know • How to internetwork(heterogenous) • IP scalability • Routers don’t need to know about every host • Need to know about all the networks • Ten of thousands of networks • What about scalability? • Routing tables don’t scale • IGP don’t scale Internetworking

31. Scaling • Two related scaling issues • Scalability of routing • Minimize # of network numbers in routing protocols • Minimize # of entries in the routing tables • Classless routing • Hierarchy • Address utilization • Minimize the usage rate if IP addresses • Classless routing Internetworking

32. Making Routing Scale • Need efficient Hierarchical Addressing • Reduce the # of network numbers assigned • Inefficient use of Hierarchical Address Space • class C with 2 hosts (2/255 = 0.78% efficient) • class B with 256 hosts (256/65535 = 0.39% efficient) 7 24 (a) 0 Network Host 14 16 Any network with > 255 hosts (b) 1 0 Network Host 21 8 (c) 1 1 0 Network Host Internetworking

33. Network number Host number Class B address 111111111111111111111111 00000000 Subnet mask (255.255.255.0) Network number Subnet ID Host ID Subnetted address Subnetting • Split a single IP network number into several subnets • Add another level to address/routing hierarchy: subnet • Subnet masksenable physical networks to share the samesubnet number • Subnets visible only within site, world sees single network Configure each node with a subnet mask Hosts now configured with IP address and subnet address Internetworking

34. Subnet mask: 255.255.255.128 Subnet number: 128.96.34.0 128.96.34.15 128.96.34.1 H1 R1 Subnet mask: 255.255.255.128 128.96.34.130 Subnet number: 128.96.34.128 128.96.34.139 128.96.34.129 H2 R2 H3 128.96.33.1 128.96.33.14 Subnet mask: 255.255.255.0 Subnet number: 128.96.33.0 Subnet Example Forwarding table at router R1 Subnet Number Subnet Mask Next Hop 128.96.34.0 255.255.255.128 interface 0 128.96.34.128 255.255.255.128 interface 1 128.96.33.0 255.255.255.0 R2 Internetworking

35. Datagram Forwarding Algorithm D = destination IP address for each entry (SubnetNum, SubnetMask, NextHop) D1 = SubnetMask & D if D1 = SubnetNum if NextHop is an interface deliver datagram directly to D else deliver datagram to NextHop • Use a default router if nothing matches • Not necessary for all 1s in subnet mask to be contiguous • Can put multiple subnets on one physical network • Subnets not visible from the rest of the Internet Internetworking