How is it different from Circuit Cross-connect?

There are two major advantages of Layer 2 MPLS VPNs over Circuit Cross-connect: First, with Circuit Cross-connect, you must configure point-to-point circuits. Building a VPN in this manner is tedious, because adding or removing a site requires a fair amount of work, as well as touching lots of routers. With Layer 2 MPLS VPNs, you can configure a VPN specifically, so the configuration is much more straightforward. Adding or removing a site is also much simpler, and in many cases it involves touching only the one router attached to the site. Second, each Circuit Cross-connect requires a dedicated provider edge-to-provider edge label-switched path (PE-to-PE LSP), and this LSP must be signalled by using

RSVP. (Using Label Distribution Protocol [LDP] makes no sense in a Circuit Cross-connect, point-to-point scenario, because it employs a reverse-path tree approach.) Layer 2 MPLS VPNs use label stacking, so a single PE-to-PE LSP can be used for multiple VPNs and, in fact, for multiple services. This makes it an extremely scalable technology. Furthermore, this LSP can be set up through any means, including RSVP and LDP. Label-stacking techniques also support greater infrastructure scalability.