Email to a friend
Print version
Comments

Protocol Independent Multicast-Sparse Mode

Protocol Independent Multicast-Sparse Mode

Overview

Protocol Independent Multicast-Sparse Mode (PIM-SM) is similar to PIM-DM in that both protocols depend on the underlying unicast routing protocol for determining RPF interfaces. A sparse mode protocol is assumed to operate in an environment where the multicast sources and multicast receivers are not closely located, so the distribution of PIM-SM nodes is sparse. This does not imply that PIM-SM cannot be used in a LAN environment but implies that sparse mode protocols operate more efficiently over Wide Area Networks (WAN). Dense mode protocols, on the other hand, use a broadcast and prune methodology, whereas multicast routers assume everyone wants to receive multicast traffic. Under this model, traffic from a multicast source is sent on all downstream interfaces until an interface is pruned from the multicast tree. An interface has a limited prune time, after which the interface is grafted back onto the multicast delivery tree and multicast traffic is again flooded onto the network.

Sparse mode protocols use an explicit join model in which multicast traffic is only forwarded onto an interface if receivers downstream have joined the group. Dense mode protocols, however, use source trees that are dynamically created for each source using the Reverse Path Forwarding (RPF) technique. PIM-SM uses shared trees for the delivery of multicast traffic. A shared tree contains a central point to which all senders of a particular multicast group send their traffic (see Figure 7-1). Each sender routes traffic along the shortest path to the central point, which then distributes the traffic to all receivers of the group along the shortest path. The group central point in PIM-SM is referred to as the Rendezvous Point (RP). Multiple RPs can exist in a network, but there should only be one RP for a particular multicast group.

Figure 7-1: PIM-Sparse Mode shared delivery tree

Figure 7-2 actually contains three source-based trees, depending on how you look at it. Assume the RP is the receiver of the multicast traffic; the paths from routers A and B are the source-based trees because the traffic flows along the shortest path given by the RPF interfaces. Now assume the RP is the sender of the multicast traffic. The path to every receiver in the group from the RP is again the shortest path tree. When these three trees are combined, you have the shared tree of PIM-SM. The combination of these trees is not necessarily the shortest path between the senders and the receivers, as can be seen in Figure 7-2. In the figure, we have the same network topology as in Figure 7-1, except now we are running PIM-DM instead of PIM-SM. Thus, two source trees follow the shortest path from each sender to each receiver.

Figure 7-2: PIM-Dense Mode source delivery trees

You may be thinking, what’s the point? Why not use the source-based trees instead of the shared tree because the shared tree is not the optimum path? This question can be answered in two ways. The first answer is that PIM-SM has a mechanism that allows the last hop router, the one with directly attached receivers, to join the source tree and leave the shared tree. This process is called shortest path tree (SPT) switchover. The decision to switchover is based on configured thresholds that we will examine later in the chapter. The second answer is sparse mode routers do not maintain as much state information as dense mode routers, making the maintenance of state more efficient.

Another question that has probably come to mind concerns the RP. How do the routers know where the RP is? A brief answer is that there are three ways for routers to know the location of the RP. The first way is to manually configure the address of the RP on each router that is running PIM-SM. The other two ways are dynamic and depend on the version of PIM-SM that is being employed in the network. PIM-SM version one has a mechanism called Auto-RP and PIM-SM version 2 uses candidate RP advertisements. We will see later how to configure all three methods. For now, we will assume that all the PIM-SM routers know the location of the RP. As with PIM-DM, the trees are constructed by using the routes in the unicast routing table. As we have seen in the previous chapter, the shared tree may not always be the same for a different unicast routing protocol.