MPLS-Enabled Applications: Emerging Developments and New Technologies

6.1. INTRODUCTION

In the Foundation chapter of this book (Chapter 1), we discussed how MPLS is used to establish LSPs in the network and how the form of the LSP depends on the signaling protocol used. We saw that when RSVP is the signaling protocol, each LSP is point to point in nature, carrying traffic from one ingress point to one egress point. In contrast, when LDP is the signaling protocol, each LSP is multipoint to point in nature, carrying traffic from several ingress points to a single egress point.

In this chapter we will see how RSVP or LDP can be used to create point-to-multipoint (P2MP) LSPs which carry traffic from one ingress point to several egress points, thus enabling multicast forwarding in an MPLS domain. Using P2MP LSPs, traffic is multicast from one source to multiple destinations in a bandwidth-efficient manner, without the ingress having to send separate copies to each receiver.

The use of RSVP-based P2MP traffic engineering gives the ingress router control over the path taken by the traffic and allows bandwidth guarantees to be made. As described later in this chapter, this unification of traffic engineering and multicast enables applications that were previously difficult to support on an IP or MPLS network, such as the distribution of broadcast-quality television.

In later chapters, we discuss the use of P2MP LSPs in the context of L3VPN and VPLS, for the transport of customers' IP multicast traffic. The P2MP LSPs are set up in the service provider's core ...