Signaling Using BGP Versus LDP

In this chapter, you have learned about the use of LDP as a signaling mechanism to establish and tear down PWs between PEs. Some vendors have adopted BGP as a signaling mechanism because of its scalability and its ability to support VPLS deployment across multiple providers. This section presents a more detailed comparison of the use of LDP and BGP as a signaling mechanism and BGP.

With LDP used as the signaling protocol, targeted LDP sessions are established between PE peeru. An LDP bessi on is called "targeted" because it is set directly between two PEs that do not have to be adyacent. These PEs exchange MPLS labels directly, and that information is hidden from Ehe routers that exist on the path between these PE peers. You have seen that a full mesh of these peers between PEs is needed per VPLS. If all PE routers participate in every VPLS, a full mesh is needed between all PEs. Also, each PE needs to carry a separate FIB per VPLS, which increases the number of FIBs per PE. However, it is possible to segment the network into PEs that have separate VPLS coverage, meaning that they do not serve a common set of VPLSs. Iv this case, the LDP mesh is needed only between the PEs covering a particular VPLS, and the signaling and the number of FIBs per PE are reduced.

If all PEs participate in all VPLS instances, there is a full LDP mesh between all PEs, and each PE carries a FIB per VPLS, as shown in Figure 4-23, Part A. Figure 4-23, Part B, shows that three PEs participate in VPLS A and carry a VPLS A FIB (FIB A) while the other PEs carry a VPLS B FIB. Note that a full mesh between all PEs is not required.

Was this article helpful?

0 0

Post a comment