Hello Ahmed,

option A can coexist with option B because option A requires back to back connections in each VRF.

Option A:

The ASBR PE needs to configure one VRF and one VRF access link for each VPN that needs to be extended to other provider.

The limit is the max. number of logical interfaces that can be created on the node (given by IDB limit it should be roughly 4096 for 7609)

see

https://www.cisco.com/en/US/products/sw/iosswrel/ps1835/products_tech_note09186a0080094322.shtml#max_no_vlans

Option B puts pressure on memory usage on PE -ASBR because they need to exchange VPNv4 prefixes and labels for each VPN to be extended.

Option C: moves the VPNv4 signalling on Route reflectors using VPNv4 eBGP sessions with multihop (and these need to be different nodes) the ASBR PEs exchange routes related to all PE loopbacks and have to build and mantain LSPs with destination the PE loopback addresses.

For effective option C you need to deploy a RRS that will have eBGP VPNv4 sessions with selected carriers.

on the c7609 option A, B can coexist with the PE looopback addresses info for option C

Hope to help

Giuseppe