[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: LC comment on "considerations" inter-AS section



Eric, working group,

Taking a bit perspective, let me try to summarize why we pursue this thread:
- the possible P-tunnel aggregation doable with the segmented inter-AS approach is currently *not* mentioned in the document (and thus not a "claim" of the document as you suggest below) - you brought up this point about P-tunnel aggregation and debate on the associated scaling benefits, this is indeed an additional potential argument for the segmented approach, and thus only tipping the balance just slightly further in favor of that approach (thus, not in the direction of changing the conclusions)

From there we have two choices:
a - do not talk about this additional potential advantage of the segmented approach b - talk about it and identify the conditions when it provides an advantage for the segmented approach and on what ASBRs need to do for this benefit to happen, and evaluate when the advantage is big or small

For sake of completeness, I guess we should go the (b) way and consolidate the different observations made to add content in section 3.5.

Other opinions are welcome.

That said, a few elements with reference to the points you brought up:
- on whether or not aggregation benefits S-PMSI: P-tunnel aggregation in the segmented approach is not only provided for I-PMSI, but also for S-PMSI, and described in section 12.2.1 of draft-ietf-2547bis-mcast-bgp - on whether or not you can avoid sending traffic uselessly in other ASes : this is mentioned in section 9.2.3.6 of draft-ietf-2547bis-mcast-bgp (the whole supersection is on I-PMSI route processing, but the text in 9.2.3.6 would apply equally well to S-PMSIs, possibly an improvement to bring there) - some other questions you ask me are answered, or should be answered in draft-ietf-2547bis-mcast and/or draft-ietf-2547bis-mcast-bgp: if they are not, the fix would be supposed to be made there, and not in draft-ietf-l3vpn-mvpn-considerations

-Thomas