Home > Resources > Multicast Distribution Trees

Multicast Distribution Trees

The MVPN feature establishes at least one multicast distribution tree (MDT) for each multicast domain. The MDT provides the information needed to interconnect the same MVRFs that exist on the different PE routers. MVPN supports two MDT types: • Default MDT—The default MDT is a permanent channel for PIM control messages and low-bandwidth streams between all PE routers in a particular multicast domain. All multicast traffic in the default MDT is replicated to every other PE router in the domain. Each PE router is logically seen as a PIM neighbor (one hop away) from every other PE router in the domain. • Data MDT—Data MDTs are optional. If enabled, they are dynamically created to provide optimal paths for high-bandwidth transmissions, such as full-motion video, that do not need to be sent to every PE router. This allows for on-demand forwarding of high-bandwidth traffic between PE routers, so as to avoid flooding every PE router with every high-bandwidth stream that might be created.

To create data MDTs, each PE router that is forwarding multicast streams to the backbone periodically examines the traffic being sent in each default MDT as follows: 1. Each PE router periodically samples the multicast traffic (approximately every 10 seconds for software switching, and 90 seconds for hardware switching) to determine whether a multicast stream has exceeded the configured threshold. (Depending on when the stream is sampled, this means that in a worst-case scenario, it could take up to 180 seconds before a high-bandwidth stream is detected.) Note Data MDTs are created only for (S, G) multicast route entries within the VRF multicast routing table. They are not created for (*, G) entries. 2. If a particular multicast stream exceeds the defined threshold, the sending PE router dynamically creates a data MDT for that particular multicast traffic. 3. The sending PE router then transmits a DATA-MDT JOIN request (which is a User Datagram Protocol (UDP) message to port 3232) to the other PE routers, informing them of the new data MDT. 4. Receiving PE routers examine their VRF routing tables to determine if they have any customers interested in receiving this data stream. If so, they use the PIM protocol to transmit a PIM JOIN message for this particular data MDT group (in the global table PIM instance) to accept the stream. Routers that do not currently have any customers for this stream still cache the information, in case any customers request it later on. 5. Three seconds after sending the DATA-MDT JOIN message, the sending PE router removes the high-bandwidth multicast stream from the default MDT and begins transmitting it over the new data MDT. 6. The sending PE router continues to send a DATA-MDT JOIN message every 60 seconds, as long as the multicast stream continues to exceed the defined threshold. If the stream falls below the threshold for more than 60 seconds, the sending PE router stops sending the DATA-MDT JOIN messages, and moves the stream back to the default MDT. 7. Receiving routers age out the cache information for the default MDT when they do not receive a DATA-MDT JOIN message for more than three minutes. Data MDTs allow for high-bandwidth sources inside the VPN while still ensuring optimal traffic forwarding in the MPLS VPN core. Note For technical information about the DATA-MDT JOIN message and other aspects of the data MDT creation and usage, see the Internet-Draft, Multicast in MPLS/BGP IP VPNs, by Eric C. Rosen et al. In the following example, a service provider has a multicast customer with offices in San Jose, New York, and Dallas. The San Jose site is transmitting a one-way multicast presentation. The service provider network supports all three sites associated with this customer, in addition to the Houston site of a different enterprise customer. The default MDT for the enterprise customer consists of provider routers P1, P2, and P3 and their associated PE routers. Although PE4 is interconnected to these other routers in the MPLS core, PE4 is associated with a different customer and is therefore not part of the default MDT. Figure 39-1 shows the situation in this network when no one outside of San Jose has joined the multicast broadcast, which means that no data is flowing along the default MDT. Each PE router maintains a PIM relationship with the other PE routers over the default MDT, as well as a PIM relationship with its directly attached PE routers.

Figure 39-1 Default Multicast Distribution Tree Overview

If an employee in New York joins the multicast session, the PE router associated for the New York site sends a join request that flows across the default MDT for the multicast domain. The PE router associated with the multicast session source (PE1) receives the request. Figure 39-2 shows how the PE router forwards the request to the CE router associated with the multicast source (CE1a).
Figure 39-2 Initializing the Data MDT

The CE router (CE1a) starts sending the multicast data to the associated PE router (PE1), which recognizes that the multicast data exceeds the bandwidth threshold at which a data MDT should be created. PE1 then creates a data MDT and sends a message to all routers using the default MDT that contains information about the data MDT. Approximately three seconds later, PE1 begins sending the multicast data for that particular stream using the data MDT. Because only PE2 has receivers who are interested in this source, only PE2 joins the data MDT and receives traffic on it.

0 Responses

Comment

Contact Us

86-136-2222-6316
CALL ME NOW

© 2011 CathaySchool, an ANDA Technology Group company, All Rights ReservedPrivacy Policy | Refund Policy | Disclaimer | Sitemap | Resources Tags