BTV multicast

Overview

This section describes how the NFM-P can be used to configure and manage the delivery of BTV multicast traffic streams. See BTV multicast configuration examples for examples.

Optimizing for broadcast TV means implementing multicast packet replication throughout the network. Multicasting improves the efficiency of the network by reducing the bandwidth and fiber needed to deliver broadcast channels to the end user. A multicasting device can receive a single copy of a broadcast channel and replicate it to any downstream devices that require it, thus substantially reducing the required network resources. This efficiency becomes increasingly important closer to the end user.

Multicast routing overview

Multicast routers direct traffic to several receivers without incurring any additional overhead for the source router or the receivers. In contrast, unicast networks can suffer from increased bandwidth requirements as the number of receivers increases.

In multicast routing, receivers query the source router to request a specific data stream. The multicast group is not restricted by physical location, so long as the hosts can be reached through the Internet. Routers in the network must use IGMP to send and receive multicast data streams.

A multicast-enabled device, such as a switch or router, distributes a data stream to multiple receivers. Multicast packets are replicated in the network by routers that are enabled with PIM, which results in the efficient delivery of data to multiple receivers using less bandwidth.

  1. A switch or router distributes a data stream to multiple receivers, such as multicast-enabled PE switches or routers.

  2. The multicast-enabled switch or router replicates the data stream, when required, and transmits a copy to each downstream switch or router in the multicast tree.

  3. Each client receives the data stream it has subscribed to from the downstream switch or router.

The NEs involved in delivering BTV multicast streams are first preconfigured through CLI for discovery and management by NFM-P. After discovery, routing protocols are applied to the NEs using NFM-P. Routing, QoS, and network queue policies are then created. On some devices, multicast package and ACL filter policies are created. These policies are applied to NEs during service creation through NFM-P.

Content delivery

BTV source traffic consists of one IP multicast stream per broadcast channel. As a multicast stream enters the core network, it is directed by PIM to the RP, which replicates the multicast traffic to all DRs that have requested the specific multicast stream. DRs distribute the multicast stream directly to set-top receivers or through an M-VPLS to BTV VLAN rings to which customer set-top receivers are connected. Multicast streams are forwarded only to those set-top receivers that have requested them through IGMP and are entitled to them as subscriber hosts.

Content management

Multicast package policies on some devices define the available multicast addresses (BTV channels) for end users in a BTV network. Typically, a root package policy which includes all BTV channels associated with an NFM-P customer service is created. Subsets of the root policy are then created as BTV content packages to which end customers can subscribe. ACL filter policies on CE devices ensure that only the channels to which an end customer has subscribed are delivered to the customer set-top receiver.

PIM

PIM uses RPF to correctly forward multicast packets down a distribution tree, using the independent multicast and unicast routing tables created by the 7450 ESS in mixed mode or the 7750 SR. The unicast routing table is populated by the unicast routing protocols, such as OSPF, BGP, IS-IS, or static routes, which can also be configured to submit routes to the multicast routing table.

Depending on the configuration of the PIM routing instance, RPF can use the unicast routing table, the multicast table populated by the unicast routing protocols, or both to determine the upstream sources of multicast streams. PIM forwards a multicast packet only if it is received on an upstream interface that is associated with a source address of an upstream router. This RPF check assures that there are no loops in the distribution tree.

PIM uses a multicast domain to group receiver hosts on a router called the rendezvous point (RP). A bootstrap router (BSR) elects an RP from available candidates. The BSR manages RP information, disseminates it to all PIM routers in the multicast domain, and elects a new RP in the case of RP unavailability.

A receiver host becomes a member of a multicast domain by sending an IGMP join request for a multicast stream to a PIM designated router (DR). If the router does not currently receive the multicast stream, PIM updates the DR routing table with the receiver host IP address and requests the multicast stream from the RP. The RP adds the router to the distribution tree. Packets sent to the multicast IP address are propagated down the distribution tree to the receiver host. DRs use the RP as the source for a multicast stream unless a source with a lower path cost is available.

PIM stops sending a multicast stream to a router when it determines that there are no active receiver hosts for the multicast stream in that branch of the distribution tree.

MVR on VPLS

PIM is not supported on 7450 ESSs. When receiver hosts are connected to a PIM DR by way of a 7450 ESS, MVR must be configured on the switch. MVR allows multicast traffic to be forwarded downstream from the DR to the receiver host over an MVR VPLS.

IGMP

IGMP is a multicast protocol which service providers can use to establish multicast group memberships on a LAN. Within the LAN, end users use IGMP to communicate with a local multicast router, which then uses PIM to distribute the IGMP messages to other local and remote multicast routers. Multicast routers send regular membership queries to IGMP hosts which respond with membership reports. Multicast routers can use these reports to determine which hosts are interested in receiving particular multicast messages.

IGMP operates above the network layer on IPv4 networks.

MLD and MLD-snooping

The Multicast Listener Discovery protocol is essentially the IPv6 version of IGMP. It is used by IPv6 routers to discover the presence of multicast listeners (that is, NEs that wish to receive multicast packets) on their directly attached links, and to discover specifically which multicast addresses are of interest to those neighboring NEs.

MLD version 2, or MLDv2, can interoperate with MLD version 1, or MLDv1. MLDv2 adds the ability for an NE to report interest in listening to packets with a specific multicast address only from specific source addresses or from all sources except for specific source addresses.

Most NFM-P NEs support MLD-snooping but not MLD. The 7450 ESS, 7750 SR, and 7950 XRS routers allow the enabling of MLD snooping for VPLS services. All variants of the 7705 SAR support MLD but not MLD-snooping.