Multicastingand multicast routing protocols

 In unicasting, the router forwards the received packet through
only one of its interfaces.
 The relationship between the source and the destination is one-to-
one.
Figure 1 Unicasting.
2

 In multicast communication, there is one source and a group of
destination.
 In multicasting, the router may forward the received packet
through several of its interfaces.
 The source address is a unicast address, but destination address
is a group address.
3

Packets are
duplicated in
routers
Figure 2 Multicasting.
4

 Emulation of multicasting through multiple unicasting is not
efficient and may create long delays, particularly with a large
group.
5
Multicasting
Multiple Unicasting

One-to-many:
 Scheduled audio-video distribution (lectures, business TV)
 Push media (news headlines, weather updates, radio broadcast)
 File distribution and caching (web site content, file-based updates)
 Announcements (network time, session schedules)
 Monitoring (stock prices, telemetry)
Many-to-many:
 Multimedia conferencing (audio/video, whiteboards)
 Synchronized resources (shared distributed databases)
 Concurrent processing (distributed parallel processing)
 Collaboration (shared document editing)
 Distance learning (one-to-many + feedback)
 Chat groups 6

 In unicast routing, each router in the domain has a table that
defines a shortest path tree to possible destinations.
7
Figure 4 Shortest Path Tree (Unicast Routing).
.

 Basic requirements of multicast routing:
 Every group member should receive only one copy of the sent packet.
 Nonmembers must not receive a copy.
 A packet must not visit router more than once (no loops).
 Paths from the source to each destination must be optimal (shortest path).
 In multicast routing, each involved router needs to construct a
shortest path tree for each group.
 Multicast trees (with source at the root and the group members
being lives) are called spanning trees. The optimal tree is called
shortest path spanning tree.
8

 In the source-based tree approach, each router needs to have one
shortest path tree for each group.
9
Figure 5 Source-based tree approach.

 In the group-shared tree , only the core router, which has a
shortest path tree for each group, is involved in multicasting.
10
Figure 6 Group-shared tree approach.

 In multicast routing, Packets from remote sources will only be
forwarded by IP routers onto a local network only if they know
there is at least one recipient for that group on that network.
 Internet Group Management Protocol (IGMP)
 Used by end hosts to signal that they want to join a specific
multicast group.
 Used by routers to discover what groups have interested
member hosts on each network to which they are attached.
 Implemented directly over IP.
11

12
Figure 7 Taxonomy of common multicast protocols.

 Multicast link state routing is a direct extension of unicast
routing, uses the source-based tree approach.
 Every router creates n (n is the number of groups) shortest path
trees, using Dijkstra’s algorithm.
 Problem with this protocol is the time and space needed to
create and store that much trees.
 MOSPF(Multicast Open Shortest Path First)
 It used multicast link state routing to create source-based trees.
 It calculates trees on demand.
13

 Multicast distance vector routing uses source-based trees, but the
router never actually makes a routing table.
 The trees are evanescent, that is after a packet is forwarded the
table is destroyed.
 To accomplish the routing, multicast distance vector algorithm
uses a process based on one of the four decision-making
strategies.
 Flooding.
 Reverse Path Forwarding (RPF).
 Reverse Path Broadcasting (RPB).
 Reverse Path Multicasting (RPM).
14

 Flooding:
 In flooding, a router receives a packet and sends it out to every
interface except the from which it was received.
 Flooding broadcasts packets not multicast.
 It creates loops in the systems.
 Reverse Path Forwarding (RPF):
 RPF eliminates the loops in the flooding process.
 In RPF , a router forwards only one copy of a packet that has
traveled the shortest path from the source to the router.
 To checkout if the packet has come though its shortest path RPF
uses unicast routing table.
15

16
Figure 8 An Example of RPF .

 RPF can eliminate loops but can not guarantee that each network
receives a single copy of a packet.
17
Figure 9 Problem with RPF

 Reverse Path Broadcasting (RPB):
 RPB creates a shortest path broadcast tree from the source to each
destination.
 It guarantees that each destination receives one and only one
copy of the packet.
 For each network a designated parent router is selected. The router will
send a packet to a network only if that router is the designated parent router
for this network.
18
Figure 10 RPF versus RPB

 Truncated RPB (TRPB):
 Both, RPF and RPB broadcast the packets. Consequently a network
that doesn't contain the multicast group will receive the packet, and the
second layer of each host in the network will decide whether to deliver
or to drop the packet, based on the MAC address. Which is very
inefficient.
 In TRPB a designated parent router can determine (via IGMP)
whether members of a given multicast group are present on the router
sub-network or not. If this sub-network is a leaf sub-network (it doesn't
have any other router connected to it) the router will truncate the
spanning tree.
19

 Reverse Path Multicasting (RPM):
 RPM adds pruning and grafting to RPB to create a multicast shortest path
tree that supports dynamic membership changes.
 Pruning:
 The router sends a prune message to the upstream router ,so that the
upstream router can stop sending multicast messages for this group
though that interface.
 Grafting:
 The graft message forces the upstream router to resume sending the
multicast messages.
20

21
Figure 11 Reverse Path Multicasting (RPM).

22
Figure 12 RPF, RPB, and RPM.

 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.
 CBT algorithm uses shared trees, instead of source-based trees
23
Rendezvous point
(core router) gets
JOIN/LEAVE
requests and forms
the shared tree
Unicast packets
(JOIN/LEAVE
requests)
Hosts A and C
want to join
group
Figure 13 CBT structure.

24
Host E sends a
message to
group G1. This
message is unicast
to the RP.
CBT routers
decapsulate the
message from E and
multicast it to the
group G1
Sender doesn’t
necessarily need
to be a member
of the group.
Anybody can be
a sender, this
won’t change the
CBT.
This router
doesn’t have to
be a CBT router
Each CBT
router has an
entry for group
G1 in routing
table
Figure 14 CBT Routing structure.

Advantages:
 Smaller routing tables (only one entry per group).
 Senders do not need to join a group to send messages.
 Simple and robust (only core router maintains connectivity).
Disadvantages:
 Shared trees are not as optimal as source-based trees.
 Core routers can become bottlenecks.
 A single point of failure.
25

 Protocol Independent Multicast (PIM) is the name given to two
independent multicast routing protocols:
 Protocol Independent Multicast, Dense Mode (PIM-DM), and
 Protocol Independent Multicast, Sparse Mode (PIM-SM).
 Dense mode (PIM-DM):
 Uses source-based trees.
 PIM-DM uses RPF and pruning/grafting strategies to handle multicasting.
 It is independent from the underlying unicast protocol.
 Used in a dense multicast environment, such as a LAN environment.
 Sparse mode (PIM-SM):
 Uses shared trees.
 PIM-SM is used in a sparse multicast environment such as a WAN.
 PIM-SM is similar to CBT but uses a simpler procedure.
26

 A multicast router may not find another multicast router in the
neighborhood to forward the multicast packet.
 A solution for this problem is tunneling. We make a multicast
backbone (MBONE) out of these isolated routers using the
concept of tunneling.
27
Virtual point-to-
point link
Isolated
island of
routers
nonmulticast
routers
Figure 15 Logical tunneling.

 Easy to deploy (no explicit router support).
 Manual tunnel creation/maintenance.
 No routing policy – single tree.
28
Figure 16 MBONE.

29
IP header
G=224.x.x.x
Data
Nonmulticast
routers
IP header
G=224.x.x.x
Data
Encapsulator
( router entry
point of
the tunnel)
Decapsulator
( router exit
point of
the tunnel)
Figure 16 MBONE IP in IP Tunneling

Multicastingand multicast routing protocols

More Related Content

Multicastingand multicast routing protocols