 | Copyright © 2008, Juniper Networks, Inc. All rights reserved. Trademark Notice. | |
Protocol Independent Multicast (PIM) version 2 supports multicast over Layer 3 virtual private networks (VPNs) based on RFC 2547, BGP/MPLS VPNs, and Section 2 (Multicast Domains) of Internet draft draft-rosen-vpn-mcast-07.txt, Multicast in MPLS/BGP IP VPNs. This implementation does not require the provider (P) routers to maintain any VPN-specific PIM information, but this lack of VPN-specific information is not optimal. The issue is that a single multicast group is defined for each VPN to carry multicast control and data traffic inside the provider core and all VPNs are mapped to this single group in the provider's space. This mapping results in the delivery of packets to each provider edge (PE) router attached to the P router even if the PE router has no receivers for traffic from a multicast group in that VPN. Each PE router must process the encapsulated VPN traffic even if the multicast packets are then dropped. This is a waste of resources, especially in environments characterized by low bandwidth links in the core or a multicast source in the VPN sending a very high volume of information (for example, high-definition television [HDTV] packets) through the core.
A data multicast distribution tree (MDT), based on section 7 of Internet draft draft-rosen-vpn-mcast-07.txt, Multicast in MPLS/BGP IP VPNs, solves the problem of P routers flooding unnecessary multicast information to PE routers that have no interested receivers for a particular VPN multicast group. The multicast data MDT solution requires the creation of a new tunnel by the PE router if the source exceeds a configured rate threshold parameter. All other PE routers join the new tunnel only if the PE router has receivers in the VPN for that multicast group.
This chapter provides the following information about data MDTs:
| Copyright © 2008, Juniper Networks, Inc. All rights reserved. Trademark Notice. | |