On 4/4/22 15:45, Masataka Ohta wrote:
MPLS with nested labels, which is claimed to scale because nesting represents route hierarchy, just does not scale because source hosts are required to provide nested labels, which means the source hosts have the current most routing table at destinations, which requires flat routing without hierarchy or on demand, that is, flow driven, look up of detailed routing tables of destinations at a distance.
This detail is limited to PE devices (ingress/egress). You don't need to carry a BGP table in the P devices (core), as only label swapping is required. Fair point, it is a little heavy for an edge box, but I imagine nearly any feature of consequence is going to be high-touch, high-impact, for the edge. Those who have solved this problem with SR can comment, as we don't run it. We did experiment with IS-IS hierarchy (L1 within the data centre and L2 between them), but Route Leaking (copying L2 routes into L1) was a requirement in order to facilitate FEC creation (/32 for IPv4, /128 for IPv6). In the end, having a flat L2 domain was just simpler. It's been years, and on today's hardware, we've never ran into an issue carrying thousands of IS-IS IPv4/IPv6 routes this way. Mark.