1. Multicasting  A message can be unicast, multicast, or broadcast. Let us clarify these terms as they relate to the Internet

2. Unicasting In unicasting, the router forwards the received packet through only one of its interfaces.

3. Multicasting

4.  In multicasting, the router may forward the received packet through several of its interfaces.  In multicasting communications there is one source and a group of destinations  In broadcasting there is one source but all of other hosts are the destinations

5. Multicasting and multiple unicasting  Multicasting starts with one single packet from the source that is duplicated by the router  The destination address in each packet is the same for all duplicates  In multiple unicasting several packets starts from the source. Destination address will be different in each packet  There may be multiple copies traveling between two routers---E.g. Email  There will be a delay between packets in MU

6. Multicasting versus multiple unicasting

7. Multicast Applications  Multicasting has many applications today such as  access to distributed databases  information dissemination  teleconferencing  distance learning

8. Multicast Routing  In multicast routing, each involved router needs to construct a shortest path tree for each group  When a router receives a multicast packet it forwards to different networks Two types  Source based tree  Group shared tree

9. Source Based Tree  In the source-based tree approach, each router needs to have one shortest path tree for each group.  If the number of groups is m, each router needs to have m shortest path trees, one for each group

10. Source-based tree approach

11. Group Shared Tree  In the group-shared tree approach, only the core router, which has a shortest path tree for each group, is involved in multicasting.  Instead of each router having m shortest path tree, only one designated router called the centre core or rendezvous router, takes the responsibility of distributing multicast traffic

12. Group-shared tree approach

13. Taxonomy of common multicast protocols

14. Mulitcast link state routing  Multicast link state routing uses the source- based tree approach.  Information about a group comes from IGMP

15. Multicast Open Shortest Path First (MOSPF)  Multicast Open Shortest Path First protocol is an extension of OSPF protocol that uses multicast link state routing to create source based tree.

16. Multicast distance vector routing  Multicast distance vector routing uses source based tree  But router never makes the routing table  When a router receives a multicast packet it forwards the packet as though it is consulting a routing table.

17. Flooding  A router receives a packet and without even looking at the destination group address, send it out from every interface except the one from which it was received.

18. Reverse path forwarding(RPF)  In RPF a router forwards only the copy that has traveled the shortest path from the source to the router.  No loops  But duplicate copies may receive

19. RPF

20. Figure 15.9 Problem with RPF

21. Reverse Path Broadcasting(RPB)

22. Figure 15.10 RPF versus RPB

23. Reverse path multicating(RPM)  RPM adds pruning and grafting to RPB to create a multicast shortest path tree that supports dynamic membership changes

24. RPF, RPB, and RPM

25. Distance Vector Multicast Routing Protocol (DVMRP)  Is an implementation of multicast distance vector routing.  It is a source based routing protocol

26. CBT  The Core-Based Tree (CBT) protocol is a group-shared protocol that uses a core as the root of the tree. The autonomous system is divided into regions and a core (center router or rendezvous router) is chosen for each region.

27.  Every router is informed of the unicast address of the selected rendezvous router  Each router then sends a unicast join message  Intermediate router extracts information such as unicast address of sender and interfaces through which it has passed  When all message received a tree is formed at rendezvous router

28. Group-shared tree with rendezvous router

29. Sending a multicast packet to the rendezvous router

30. In CBT, the source sends the multicast packet (encapsulated in a unicast packet) to the core router. The core router decapsulates the packet and forwards it to all interested interfaces

31.  If router wants to leave the group it sends a leave message to upstream router.  Difference  1) tree is first made and then pruned but in CBT initial no tree, joining gradually makes the tree  2) made from the root up but CBT formed from the leaves down

32.  Host Configuration: BOOTP and DHCP

33. BOOTP  The Bootstrap Protocol (BOOTP) is a client/server protocol that configures a diskless computer or a computer that is booted for the first time.  BOOTP provides the IP address, subnet mask, the address of a default router, and the address of a name server.  Configuration file

34.  RARP provides only IP address  Both client and server should be on same network

35. Operations  BOOTP server issues a passive open command on UDP port number 67 and waits for client  A BOOTP client issues an active open command on port number 68  This message is encapsulated in UDP datagram and in turn encapsulated in IP datagram  IP addresses of client and server will be all zeros and all ones  Server responds

36. Client and server on the same network

37. Client and server on two different networks

38. Use of UDP ports

39. Using TFTP  In the reply message server defines the path name of a file in which the client can find complete booting information.  Client then uses TFTP to obtain the rest of information

40. Error Control  BOOTP requires UDP uses the checksum  BOOTP client uses timers and a retransmission policy if it does not receive the BOOTP reply to a request.  Timers will be set randomly

41. BOOTP packet format

42.  Operation Code: 8 bit: 1 request 2 reply  Hardware type: type of physical network for Ethernet value is 1  Hardware Length: 8 bit: length of physical address  Hop count: 8bit Specifies max no of hops a packet can travel  Transaction ID: 4 byte: integer  Number of seconds: 16 bit: elapsed since the client started to boot  Client IP address : 4 byte  Your IP address : 4 byte

43.  Server IP address : 4 byte  Gateway IP address: 4 byte: IP address of router  Client hardware address: Physical address of client  Server name: optional 64 byte : contains domain name of the server  Bootfile name: optional 128 byte: full path name of the bootfile  Options: 64 byte: only in reply it is used server uses a number called magic cookie in the format of IP address with the value: 99.130.83.99

44. Option format

45. Options for BOOTP

46. DHCP  In BOOTP binding is predetermined  The Dynamic Host Configuration Protocol (DHCP) provides static and dynamic address allocation that can be manual or automatic

47. Static Address allocation Backward compatible with BOOTP A DHCP server has a database that statically binds physical address to IP address

48. Dynamic Address Allocation  Use a second database  When a DHCP client requests a temporary IP address, the DHCP server goes to the pool of available(unused) IP address and assigns an IP address for a negotiable period of time.  First check the static database

49. Manual and automatic configuration  Mapping the IP address to physical address configuration  In BOOTP it is manual  In DHCP it has both manual and automatic

50. Packet Format

51. DHCP packet

52. Flag  1 bit flag – to let the client specify a forced broadcast reply from the server

53. Options  Tag 53

54. Options  Other option include 51 :Lease time 58 : Renewal (T1) time value 59 : Rebinding (T2) time value

55. Transition States  DHCP client transitions from one state to another depending on the message it receives of sends

56. DHCP transition diagram

57. Initializing state  When the DHCP client first starts, it is in the initializing state.  Client broadcast the DHCPDISCOVER message  Using port 67

58. Selecting State  After sending DHCPDISCOVER message client goes to selecting state  Servers respond with DHCPOFFER  Offers IP address, Lease time  Default lease time is 1hour  Client select the offer of one server and send DHCPREQUEST message  Client goes to requesting state

59. Requesting State  The client remains in the requesting state until it receives a DHCPACK message from the server which creates a binding between the client physical address and its IP address.  After the receipt of DHCPACK client goes to bound state

60. Bound State  Client uses the IP address until the lease expires  When 50% of the lease period is reached, client sends another DHCPREQUEST to ask for renewal.  Then goes to renewal state

61. Renewal state  Client remains in the renewal state until one of two events happens 1) It can receive a DHCPACK, which renews the lease agreement Client reset the timer goes back to bound state

62. 2) No DHCPACK is received and 87.5 % of the lease time expires, the client goes to rebinding state

63. Rebinding state  The client remains in rebinding state until of three events happens 1) Client receives a DHCPNACK 2) The lease time expires 3) DHCPACK In the first two it goes to initializing state and try to get another IP address In the third it goes to bound state

64. Exchanging messages