When developing an MVPN design, which performance and scalability consideration must be taken into account?

In MVPN, PE routers maintain PIM adjacencies with their directly attached CEs; CE routers do not establish end-to-end PIM adjacencies across the MPLS core, so this is not a relevant scalability concern.

The default MDT is a shared multicast tree that all PE routers participating in a given VPN join, so any multicast data forwarded on it is replicated and delivered to every PE regardless of whether downstream CEs have active receivers. As the number of PE routers or active multicast streams grows, this unnecessary replication consumes significant core bandwidth and PE processing resources, making it the primary performance and scalability concern when designing an MVPN.

RP placement within the multicast VRF is a design consideration for intra-VPN rendezvous, but it does not drive MVPN infrastructure scalability the way default MDT replication does.

Customer network RP placement affects the customer multicast domain but does not directly impact the scalability of the MVPN provider infrastructure.