| Hi Dean, Many thanks for your comments. Replies/Discussion in line, - No topology or resource information is distributed between domains (as mandated per [RFC4105] and [RFC4216]), which is critical to preserve IGP/BGP scalability and confidentiality in the case of TE LSPs spanning multiple domains. DC2 - This is practically true but if not, the BRPC procedure can still be used. JP>> Yes, but we want to clarify the fact BRPC does not require any topology or resource information distribution across domains. We'll reword to clarify. - While certain constraints like bandwidth can be used across different domains, certain other TE constraints like resource affinity, color, metric, etc. as listed in [RFC2702] could be translated at domain boundaries. If required, it is assumed that, at the domain boundary LSRs, there will exist some sort of local mapping based on offline policy agreement, in order to translate such constraints across domain boundaries during the inter-PCE communication process. DC3 - This may also be relaxed as an assumption. These constraints may be globally or locally (thus may need mapping) defined, but in either case, the BRPC procedure may be applied. JP>> Well at least not for the cost metric that must be consistent to compute the shortest inter-domain constrained path. You're right that other metrics apply to the general inter-domain case and are not specific to BRPC. - Each AS can be made of several IGP areas. The path computation procedure described in this document applies to the case of a single AS made of multiple IGP areas, multiples ASs made of a single IGP area or any combination of the above. For the sake of simplicity, each AS will be considered to be comprised of a single area in this document. The case of an Inter-AS TE LSP spanning multiple ASs where some of those ASs are themselves made of multiple IGP areas can be easily derived from this case by applying the BRPC procedure described in this document, recursively. DC4 - The BRPC procedure in general can be applied to any network partitions in the context of (G-)MPLS networks, and so this paragraph could be moved to the Intruduction, and at least not a hard assumption. JP>> Agree. - The domain path (set of domains traversed to reach the destination domain) is either administratively pre-determined or discovered by some means (outside of the scope of this document). DC5 - What does this mean ? Thought the domain path is calculated by the BRPC procedure. JP>> nope, the BRPC computes the optimal (shortest) path between a source S in Domain A to destination D in Domain B, along a determined domain path (can be pre-determined or discovered during the BRPC procedure). But, BRPC does not attempt to discover all the domain-path, compute for each domain path the best path and return the absolute best path. DC10 Thr BRPC procedure may return x number of path segments where x equals the number of equal cost paths (ECMP), where two parameters may be configurable based on carriers policy. The first is the max number for x, and the second is the approximates of these paths (costs are exactly the same or close enough). JP> Do you see a strong reason for limiting x ? The second requirement might indeed be interesting. WG, any feed-back on this ? DC11 - It could be reworded to say: After accommodating local policy that may be associated with domains that a LSP may traverse, if any, the BRPC would guarantee the optimality of inter-domain paths end-to-end. JP> Not quite sure to see your point here ? Thanks for your comments. JP. On May 8, 2006, at 5:57 PM, Dean Cheng ((dcheng)) wrote:
|
_______________________________________________ Pce mailing list [email protected] https://www1.ietf.org/mailman/listinfo/pce
