Exercises to Communication Systems IP Multicast Dr.-Ing. Falko Dressler Kai-Steffen Hielscher Department of Computer Science 7 University of Erlangen
Group Communication Multiple partners communicate in a closed group Types of group communication Unicast: 1:1 Concast: m:1 Multicast: 1:m Multipeer: m:n (typically emulated using multicast) Other types of communication Broadcast Anycast Scalability Group size Topology Dynamics Multicast 2
IP Multicast Broadcast Routing Multicast Routing Principles Building Multicast Trees DVMRP PIM SM and DM Multicast 3
Broadcast Bradcast: act of sending datagram to all receivers with single transmit operation analogy: TV Question: how to achieve broadcast? Multicast 4
Flooding the network Uncontrolled flooding eventually will shut down the network Controlled flooding sequence-number-controlled flooding source and broadcast sequence number are stored on each node if a broadcast packet is on the stored list, it is dropped reverse path forwarding Multicast 5
Broadcast Routing Algorithms duplicate R1 duplicate creation/transmission R1 R2 R2 duplicate R3 R4 R3 R4 source duplication in-network duplication Source-duplication vs. in-network duplication. Multicast 6
Reverse path forwarding A forward packet to all neighbors F C E B D if a packet comes from the node which is the first node on the path towards the source G drop packet otherwise Multicast 7
Spanning-tree broadcast spanning tree each and every node in a graph no cycles A A Any node can be the source C B C B F E D F E D Broadcast initiated at A G G Broadcast initiated at D minimum spanning tree if the cost of all links is minimal Multicast 8
Constructing a spanning tree numerous algorithms have been developed here: center-based construction select a center point (rendezvous point - RP)»how? active research; e.g., select any send tree-join message to the center from each node (unicast) either gets to the center or gets to an existing branch F C 1 A 3 4 E 2 B D 5 Stepwise construction of spanning tree with E as an RP) G F C A E B Constructed spanning tree D G Multicast 9
IP Multicast Broadcast Routing Multicast Routing Principles Building Multicast Trees DVMRP PIM DM and SM Multicast 10
Principles of Multicast Source Source n point-to-point connections single multicast connection Unicast receivers Multicast receivers Multicast 11
Typical Scenarios: one-to-many TV broadcast time synchronization (NTP) distribution of data, e.g., stock exchange rates Receiver Receiver Receiver Receiver Sender Receiver Sender Receiver Receiver Receiver Multicast 12
Typical Scenarios: many-to-many video conferences multiplayer games Participant Participant Participant Participant Participant 5 5 5 5 Participant 4 4 4 4 5 5 5 5 4 4 5 Participant 5 5 5 Participant 5 1 1 5 Participant 2 2 2 2 5 1 1 5 2 2 5 Participant 5 1 1 Number of flows on the link Unicast Participant Participant Multicast Multicast 13
Multicast: one sender to many receivers Multicast: act of sending datagram to multiple receivers with single transmit operation analogy: one teacher to many students Question: how to achieve multicast Multicast via unicast source sends N unicast datagrams, one addressed to each of N receivers same as in broadcast via unicast routers forward unicast datagrams multicast receiver (red) not a multicast receiver (red) Multicast 14
Multicast: one sender to many receivers Multicast: act of sending datagram to multiple receivers with single transmit operation analogy: one teacher to many students Question: how to achieve multicast Network multicast Router actively participate in multicast, making copies of packets as needed and forwarding towards multicast receivers Multicast routers (red) duplicate and forward multicast datagrams Multicast 15
Multicast: one sender to many receivers Multicast: act of sending datagram to multiple receivers with single transmit operation analogy: one teacher to many students Question: how to achieve multicast Application-layer multicast end systems involved in multicast copy and forward unicast datagrams among themselves Multicast 16
Internet Multicast Service Model 128.59.16.12 128.119.40.186 multicast group 226.17.30.197 128.34.108.63 128.34.108.60 multicast group concept: use of indirection hosts addresses IP datagram to multicast group routers forward multicast datagrams to hosts that have joined that multicast group Multicast 17
Aspects of Group Communication Addressing IP Multicast addresses Group maintenance Internet Group Management Protocol Layer 2 Routing Distribution trees Routing protocols Multicast 18
Multicast groups class D Internet addresses reserved for multicast: host group semantics: o anyone can join (receive) multicast group o anyone can send to multicast group o no network-layer identification to hosts of members needed: infrastructure to deliver mcast-addressed datagrams to all hosts that have joined that multicast group Multicast 19
IP Multicast Addressing Multicast addresses = Class D addresses address range: 224.0.0.0/4 only for destination address source address is still the unicast source address Link-Local multicast addresses only available in the subnet (will not be forwarded) address range: 224.0.0.0/24 reserved addresses (examples): 224.0.0.1 - all systems 224.0.0.2 - all routers 224.0.0.5 - OSPF routers 224.0.0.6 - OSPF designated routers Multicast 20
Address Scoping Globally scoped addresses 224.0.1.0-238.255.255.255 are to be used globally in the internet Source-specific multicast 232.0.0.0/8 GLOP addresses (RFC2770) 233.0.0.0/8 reserved for statically defined addresses by organizations that already have an AS number reserved address: 233.<AS>.0/24 Administratively scoped addresses 239.0.0.0/8 like RFC1918 addresses for local use only not routed in the internet Multicast 21
TTL Thresholds same principle as in IP unicast thresholds are also used to limit multicast traffic to a particular region TTL scope Local net Site Region World Initial TTL value 1 15 63 127 TTL threshold - 16 64 128 Multicast 22
Joining a mcast group: two-step process local: host informs local mcast router of desire to join group: IGMP (Internet Group Management Protocol) wide area: local router interacts with other routers to receive mcast datagram flow many protocols (e.g., DVMRP, MOSPF, PIM) IGMP IGMP wide-area multicast routing IGMP Multicast 23
IGMP: Internet Group Management Protocol host: sends IGMP report when application joins mcast group IP_ADD_MEMBERSHIP socket option in the C socket library host need not explicitly unjoin group when leaving router: sends IGMP query at regular intervals host belonging to a mcast group must reply to query query report Multicast 24
IGMP IGMP version 1 router: Host Membership Query msg broadcast on LAN to all hosts host: Host Membership Report msg to indicate group membership randomized delay before responding implicit leave via no reply to Query RFC 1112 IGMP v2: additions include group-specific Query Leave Group msg last host replying to Query can send explicit Leave Group msg router performs group-specific query to see if any hosts left in group RFC 2236 IGMP v3: under development as Internet draft Multicast 25
Sample multicast receiver // Import some needed classes import sun.net.*; import java.net.*; int port = 5555; // Port to listen to String mcastgroup = "225.4.5.6"; // Multicast address // Create a socket and bind it to the port MulticastSocket socket = new MulticastSocket(port); // join the multicast group socket.joingroup(inetaddress.getbyname(mcastgroup)); // Create a DatagramPacket and a receive data byte buffer[] = new byte[1024]; DatagramPacket packet = new DatagramPacket(buffer, buffer.length); socket.receive(packet); // Print the received data System.out.println("Received data from: + packet.getaddress().tostring() + ":" + packet.getport() + " with length: " + packet.getlength()); System.out.println(new String(packet.getData())); System.out.println(); // Leave the multicast group and close the socket socket.leavegroup(inetaddress.getbyname(mcastgroup)); socket.close(); Multicast 26
Sample multicast sender import sun.net.*; // Import some needed classes import java.net.*; int port = 5555; // Port to send to String mcastgroup = "225.4.5.6"; // Multicast address // Create a socket MulticastSocket socket = new MulticastSocket(); // We are only sending data and not receiving, // so we don't have to join the multicast group String hi = Hello! // Create a DatagramPacket DatagramPacket packet = new DatagramPacket(hi.getBytes(), hi.length(), InetAddress.getByName(mcastGroup), port); s.send(packet, timetolive); // Send the packet s.close(); // Close the socekt when finished Multicast 27
Layer 2: Multicast Ethernet Addresses Multicast IP addresses mapped to special Ethernet MAC addresses Octet 0 Octet 1 Octet 2 Octet 3 Octet 4 Octet 5 xxxxxxx1 xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx xxxxxxxx broadcast / multicast bit Multicast 28
IP Multicast to Ethernet Address Mapping Why only 23 bit? In the early 90s Steve Deering tried to get 16 OUIs from the IEEE but could not pay for it. Any problems? 32 IP multicast addresses can be mapped to a single ethernet address. This may lead to performance problems! 32 Bit 28 Bit IP multicast address Header: 1110 239.255.0.1 Ethernet multicast address Header: 01-00-5e 01-00-5e-7f-00-01 23 Bit 48 Bit Multicast 29
Layer-2 Multicast Mechanisms Normal case: multicast = broadcast, i.e. flooding trough the LAN IGMP snooping Intelligent switches process all multicast packets, look for IGMP messages and analyze them Prerequisite for a broad use: layer-3-aware switches Cisco Group Management Protocol (CGMP) Intelligence only at the router, which informs its local switches Multicast 30
CGMP Cisco Group Management Protocol Messages via the well-known CGMP MAC multicast address 0x0100.0cdd.dddd The router processes the information for the switch(es) No processing power at the switch is required Multicast 31
CGMP Messages GDA USA Join/Leave Meaning (Group Destination Address) (Unicast Source Address) Mcst MAC Client MAC Join Add port to group Mcst MAC Client MAC Leave Delete port from group 0000...0000 Router MAC Join Assign router port 0000...0000 Router MAC Leave Deassign router port Mact MAC 0000...0000 Leave Delete group 0000...0000 0000...0000 Leave Delete all grouos Multicast 32
Multicast Routing: Problem Statement Goal: find a tree (or trees) connecting routers having local mcast group members tree: not all paths between routers used source-based: different tree from each sender to rcvrs shared-tree: same tree used by all group members Shared tree Source-based trees Multicast 33
Approaches for building mcast trees Approaches: source-based tree: one tree per source shortest path trees reverse path forwarding group-shared tree: group uses one tree minimal spanning (Steiner) center-based trees we first look at basic approaches, then specific protocols adopting these approaches Multicast 34
Shortest Path Tree mcast forwarding tree: tree of shortest path routes from source to all receivers Dijkstra s algorithm S: source R1 1 R2 3 4 2 R4 5 R5 LEGEND router with attached group member router with no attached group member R3 R6 6 R7 i link used for forwarding, i indicates order link added by algorithm Multicast 35
Reverse Path Forwarding relies on router s knowledge of unicast shortest path from it to sender each router has simple forwarding behavior: if (mcast datagram received on incoming link on shortest path back to center) then flood datagram onto all outgoing links else ignore datagram Multicast 36
Reverse Path Forwarding: example S: source R2 R1 R4 LEGEND router with attached group member R5 router with no attached group member R3 R6 R7 datagram will be forwarded datagram will not be forwarded result is a source-specific reverse SPT may be a bad choice with asymmetric links Multicast 37
Reverse Path Forwarding: pruning forwarding tree contains subtrees with no mcast group members no need to forward datagrams down subtree prune msgs sent upstream by router with no downstream group members S: source LEGEND R1 R4 router with attached group member R3 R2 R6 P P R7 R5 P router with no attached group member prune message links with multicast forwarding Multicast 38
Shared-Tree: Steiner Tree Steiner Tree: minimum cost tree connecting all routers with attached group members problem is NP-complete excellent heuristics exists not used in practice: computational complexity information about entire network needed monolithic: rerun whenever a router needs to join/leave Multicast 39
Center-based trees single delivery tree shared by all one router identified as center of tree to join: edge router sends unicast join-msg addressed to center router join-msg processed by intermediate routers and forwarded towards center join-msg either hits existing tree branch for this center, or arrives at center path taken by join-msg becomes new branch of tree for this router Multicast 40
Center-based trees: an example Suppose R6 chosen as center: LEGEND R1 3 R4 router with attached group member R3 R2 1 R6 2 R7 R5 1 router with no attached group member path order in which join messages generated Multicast 41
Internet Multicasting Routing: DVMRP DVMRP: distance vector multicast routing protocol RFC1075 flood and prune: reverse path forwarding, source-based tree RPF tree based on DVMRP s own routing tables constructed by communicating DVMRP routers no assumptions about underlying unicast initial datagram to mcast group flooded everywhere via RPF routers not wanting group: send upstream prune msgs Multicast 42
DVMRP: continued soft state: DVMRP router periodically (1 min.) forgets branches are pruned: mcast data again flows down unpruned branch downstream router: reprune or else continue to receive data routers can quickly regraft to tree following IGMP join at leaf odds and ends commonly implemented in commercial routers Mbone routing done using DVMRP Multicast 43
Tunneling Q: How to connect islands of multicast routers in a sea of unicast routers? physical topology logical topology mcast datagram encapsulated inside normal (non-multicastaddressed) datagram normal IP datagram sent thru tunnel via regular IP unicast to receiving mcast router receiving mcast router unencapsulates to get mcast datagram Multicast 44
PIM: Protocol Independent Multicast not dependent on any specific underlying unicast routing algorithm (works with all) two different multicast distribution scenarios : Dense: group members densely packed, in close proximity. bandwidth more plentiful Sparse: # networks with group members small wrt # interconnected networks group members widely dispersed bandwidth not plentiful Multicast 45
Consequences of Sparse-Dense Dichotomy: Dense group membership by routers assumed until routers explicitly prune data-driven construction on mcast tree (e.g., RPF) bandwidth and non-grouprouter processing profligate Sparse: no membership until routers explicitly join receiver- driven construction of mcast tree (e.g., centerbased) bandwidth and non-grouprouter processing conservative Multicast 46
PIM- Dense Mode flood-and-prune RPF, similar to DVMRP but underlying unicast protocol provides RPF info for incoming datagram less complicated (less efficient) downstream flood than DVMRP reduces reliance on underlying routing algorithm no list of PIM neighbors (receivers from the same router) request for pruning sent to all PIM local routers ( neighbors ) response to request to prune from any receiver delayed 3 seconds, so other PIM routers can override (if they want to receive packets) has protocol mechanism for router to detect it is a leafnode router Multicast 47
PIM - Sparse Mode center-based approach router sends join msg to rendezvous point (RP) intermediate routers update state and forward join after joining via RP, router can switch to source-specific tree increased performance: less concentration, shorter paths R3 R2 R1 join join all data multicast from rendezvous point R6 join R4 R5 R7 rendezvous point Multicast 48
PIM - Sparse Mode sender(s): unicast data to RP, which distributes down RProoted tree RP can extend mcast tree upstream to source the source need not to be a receiver! RP can send stop msg if no attached receivers no one is listening! R3 R2 R1 join join all data multicast from rendezvous point R6 join R4 R5 R7 rendezvous point Multicast 49