PIM is a routing protocol that have the right to be provided to front multicast traffic. PIM operates separately of any certain IP routing protocol. Therefore, PIM uses the IP unicast routing table and also does not store a separate multicast routing table. (The unicast routing table is chin routing protocol-independent due to the fact that one or more routing protocols deserve to be offered to populate a solitary table.)

PIM have the right to operate in two modes, depending upon the thickness of the recipients in a multicast group. Cisco has emerged a third hybrid mode as well. The PIM settings are together follows:

• PIM thick mode (PIM-DM)— Multicast routers assume that multicast recipients are located everywhere, top top every router and also every router interface. After ~ a tree is built, the branches room pruned if a multicast team has no energetic recipients.

You are watching: Which mode of pim uses a source distribution tree?

• PIM sparse setting (PIM-SM)— Multicast routers construct a distribution tree by including branches only as recipients join a multicast group.

• PIM sparse-dense mode— Multicast routers operate in dense or sparse mode, depending on how the multicast team is configured.

In addition, 2 versions the the PIM protocol deserve to be supplied in a network: PIM variation 1 and PIM version 2.

Cisco firewalls to run ASA 7.0 or later, and also FWSM 3.1(1) or later, deserve to operate just in PIM sparse mode, return they have the right to coexist with other routers running PIM-SM or PIM sparsedense mode.

PIM thin Mode

PIM sparse setting takes a "bottom-up" method to building a multicast circulation tree. The tree is developed by start with the recipients or team members at the finish leaf nodes and also extending ago toward a central root point.

Sparse mode additionally works top top the idea of a common tree structure, where the root is not necessarily the multicast source. Instead, the source is a PIM-SM router the is centrally situated in the network. This source router is dubbed the rendezvous allude (RP).

The tree native the RP to the team members is in reality a subset the the tree that could be drawn from the resource to the group members. If a multicast source anywhere in the network deserve to register for group membership through the RP, the tree have the right to be completed end-to-end. Since of this, the sparse setting tree is dubbed a mutual tree.

Note

Sparse setting multicast flows room designated by a (source,destination) pair. The letters S and also G represent a specific resource and group, respectively. An asterisk (*) can also be offered to represent any resource or destination. Because that example, multicast flows over the shared tree are explained as (*,G) due to the fact that the mutual tree permits any source to send to a team G.

In PIM-SM, the shared tree is constructed using the following simple sequence that steps:

1. A recipient organize joins a multicast group by sending out an IGMP membership Report come the local router.

2. The router to add an (*,G) entry in its own multicast routing table, where G to represent the group IP address. The router also maintains a list of outbound interfaces where team recipients are located.

3. The router sends out a PIM join request for (*,G) toward the RP in ~ the tree"s root.

4. The surrounding PIM router obtain the sign up with request, add to a (*,G) entry in its own table, and adds the arriving interface come its perform of outbound interfaces for the group. The surrounding PIM router climate relays the sign up with request towards the RP.

5. When the RP ultimately receives the PIM sign up with request, it too adds a (*,G) entry and the showing up interface come its very own table. The shared tree has now been constructed from a recipient host to the RP.

For example, take into consideration the network shown in figure 3-16. A firewall off a public and private network and also acts as the RP because that PIM multicast routing. Three receivers (end-user hosts) in the network sign up with a solitary multicast group in ready to obtain traffic indigenous a multicast source.

Figure 3-16. A Sample Network with PIM Multicast Routers

*

Figure 3-17 illustrates the group membership process. ~ above the left side of the figure, the multicast receivers X, Y, and also Z every send one IGMP membership request to join group resolve 239.0.0.1. Router B receive the request from recipient X and likewise creates a multicast course entry (*,239.0.0.1) the points back toward the receiver. Router B also sends a PIM sign up with message because that (*,239.0.0.1) toward the RP, which adds the link in between it and the firewall to the multicast tree.

Figure 3-17 illustrates the group membership process. ~ above the left next of the figure, the multicast receivers X, Y, and also Z every send one IGMP membership request to sign up with group deal with 239.0.0.1. Router B receive the inquiry from recipient X and additionally creates a multicast course entry (*,239.0.0.1) the points back toward the receiver. Router B additionally sends a PIM sign up with message for (*,239.0.0.1) toward the RP, which adds the link in between it and the firewall come the multicast tree.

Figure 3-17. Structure a common Multicast Tree through PIM

*

Router C takes similar steps because that the IGMP request it receives from Receiver Y. Recipient Z is a slightly different case; the firewall obtain its IGMP inquiry directly since it is directly connected. The firewall adds a (*,239.0.0.1) multicast route entry to its table, pointing ago toward the receiver on the within interface.

Notice how all the IGMP membership reports terminate at the closestly router (or firewall) if PIM sign up with messages travel from router to router. After ~ the RP receives every the sign up with messages, the multicast tree is complete, as displayed in the right section of figure 3-17. The network topology has actually been redrawn slightly to present how the RP (firewall) is in ~ the root of the tree. This is dubbed a PIM mutual tree due to the fact that it is used by every the gadgets participating in the multicast group. Routers that have actually no active multicast team receivers (Router A, for example) execute not send a PIM sign up with message, therefore they execute not become component of the tree.

A shared tree always begins v the RP in ~ the root and also progresses downward towards the sheet nodes, wherein the receivers are located. Just the PIM routers room shown, because they in reality build and also use the tree. PIM mutual trees are always unidirectional. Multicast packets have the right to only begin at the RP and also be sent out toward the receivers.

Finally, a multicast source must also join the group so that traffic can circulation toward the receivers. The left section of number 3-18 illustrates this process, where the resource is linked to Router C.

Figure 3-18. Adding a Multicast resource to PIM tree

Figure 3-18. Adding a Multicast source to PIM trees

*

When a source joins a multicast group, the following steps take place:

1. A resource S starts sending website traffic to the multicast group address (239.0.0.1 in the example). Approximately this point, the multicast tree has actually not been expanded to the source. In fact, an alert that the source is sending out traffic upstream toward the RP! In the unidirectional mutual tree, this is no allowed. This point is faced in the next few steps.

2. The nearest PIM router obtain the traffic destined because that the multicast group and also realizes that it is coming from a source. The router have to register the resource with the RP so that it can become a part of the tree. The multicast packets room encapsulated in PIM register messages that are sent to the RP as unicasts.

3. The RP unencapsulates the it is registered messages and also sends the initial multicast packets under the tree toward the receivers.

The RP likewise sends one (S,G) PIM join message downstream toward the resource address so the a tree can be developed from the source to the RP. In the example, this is a (S,239.0.0.1) multicast flow. The idea is to construct a course to carry multicast data indigenous the resource to the tree"s source (the RP) so that it can circulation downward towards the receivers.

Note

The tree developed from the source to the RP is not a part of the PIM common tree. Instead, the is dubbed a shortest course tree (SPT) because it complies with a path from a router (the RP in this case) straight to the source. Because the SPT is separate from the common tree, multicast packets deserve to travel upward towards the RP there is no interfering with packets travel downward from the RP towards the receivers.

In effect, these are two unidirectional trees with the RP constantly serving together the root. 4. After the SPT has been built from the resource to the RP, there is no need to keep encapsulating the source data as register messages. The RP sends out a PIM it is registered Stop post toward the source. When the sheet node router in ~ the resource receives this, the stops sending the register messages and begins making use of the brand-new SPT path.

The right part of figure 3-18 illustrates the result tree structures. The hard arrows display the PIM mutual tree, from the RP under to the routers where receivers space located. The broken-line arrows represent the SPT the is constructed from the resource up come the RP.

Although it is not shown in this example, last-hop PIM routers are enabled to do an SPT switchover to effort to develop a more direct course to the multicast source. This procedure is very similar to the steps explained previously, where certain (S,G) flows are added to the PIM routers along the path. After an SPT switchover occurs, the RP is no longer required to be in ~ the source of the tree if a far better path can be found.

To leveling the tree structure and also improve efficiency, PIM can also support a bidirectional mode. If every PIM router sustaining a multicast group is configured for bidirectional mode, a solitary multicast tree is developed to affix the multicast resource to all its receivers.

Multicast packets can flow up or down the tree as important to disperse in the network. The PIM routers take it on designated forwarder (DF) roles, deciding even if it is to forward multicast packets top top a network segment in the appropriate direction. Since a single bidirectional tree is used, the multicast source can join the team without the PIM resource registration process.

PIM RP Designation

In PIM thin mode, every PIM router must understand the RP"s identification (IP address). After ~ all, every router has to send PIM Join/Prune messages toward the RP by using its unicast routing table to find the correct interface.

The simplest an approach of identify the RP is to manually configure its deal with in every PIM router. If there are not countless PIM routers come configure, this an approach is an extremely straightforward. However, if there are plenty of PIM routers or if the RP deal with is likely to adjust in the future, manual configuration deserve to be cumbersome.

Note

Beginning v ASA 7.0 and also FWSM 3.1(1), revolution RP configuration is the only option available. Other much more dynamic RP exploration methods are described in this section since they could be provided on PIM routers in her network.

Cisco also provides a proprietary method to automatically inform PIM-SM routers that the appropriate RP because that a group. This is known as Auto-RP. Routers that deserve to potentially become an RP space configured together candidate RPs. These routers advertise their capability over the Cisco-RP-Announce multicast deal with 224.0.1.39.

These announcements space picked up by one or more centrally located and well-connected routers that have been configured to duty as mapping agents. A mapping agent collects and also sends RP-to-group mapping info to every PIM routers over the Cisco-RP-Discovery multicast deal with 224.0.1.40.

A mapping agent deserve to limit the scope of its RP discovery information by setup the time-to-live (TTL) value in that messages. This borders how countless router hops away the info will still it is in valid. Any type of PIM router within this space dynamically learn of the candidate RPs that are available to use.

The second version of PIM also includes a dynamic RP-to-group mapping advertising mechanism. This is recognized as the bootstrap router an approach and is standards-based.

PIMv2 is comparable to the Cisco Auto-RP method. First, a bootstrap router (BSR) is identified; this router learns about RP candidates because that a group and advertises them come PIM routers. Only the BSR and candidate RPs need to be configured; all various other PIM routers learn of the suitable RP dynamically native the BSR.

These bootstrap messages permeate the entire PIM domain. The border of the advertisements can be limited by specifying PIMv2 border routers, which perform not front the bootstrap message further.

Note

If Auto-RP is being used in your network, be mindful that one ASA or FWSM firewall cannot take part in the Auto-RP process. The firewall must have actually the PIM RP deal with statically configured.

See more: Secret Pilots :: Into The Breach Secret Pilots, Unlocking Secret Pilots

However, the candidate RP announcements end 224.0.1.39 and also the Router discovery messages over 224.0.1.40 can pass with the firewall to with PIM routers on the various other side. Therefore, the Auto-RP mechanism can still work throughout the firewall, however the firewall can not directly benefit from the dynamic RP discovery itself.