Many thanks, Leonard. Yes, the latest revision addresses my concerns. Thanks! Carlos. > On Aug 19, 2024, at 5:01 PM, Leonard Giuliano <lenny@xxxxxxxxxxx> wrote: > > Carlos, > > Thank you for the thoughtful review. We have updated the draft based on > your feedback. Please let us know if the latest revision addresses your > concerns. A diff can be found here: > > https://author-tools.ietf.org/iddiff?url1=draft-ietf-mops-treedn-05&url2=draft-ietf-mops-treedn-06&difftype=--html > > Further comments inline: > > On Sat, 20 Jul 2024, Carlos Pignataro via Datatracker wrote: > > | Reviewer: Carlos Pignataro > | Review result: Almost Ready > | > | Hi! > | > | Review of draft-ietf-mops-treedn > | Type Telechat Review > | Team Internet Area Directorate (intdir) > | Reviewer Carlos Pignataro > | > | This document describes TreeDN, a tree-based CDN architecture designed to > | address scaling challenges of live streaming to mass audiences. This is > | achieved through Source-Specific Multicast (SSM), including PIM-SSM, and > | tunneling or overlays based on Automatic Multicast Tunneling (AMT). > | > | The overall architecture and document seem clear and largely complete. At times > | it might read a bit marketing-ish (e.g., S3, is this solving for all > | multicast?), though the technical details are comprehensive. This is an > | Informational-targeting document. > | > | >From a document clarity perspective, there's an opportunity to earlier on > | summarize what the architecture is. Interestingly, IDnits flags that there is > | no "Introduction" section. While I do not believe an Introduction is mandatory, > | early text describing TreeDN will help. Specifically, something like: > | "TreeDN is essentially the synthesis of SSM plus overlay networking > | technologies like AMT to deliver traffic to receiving hosts on > | unicast-only networks, or directly natively on-network when there's support > | for support multicast." > > Added a summarization of the solution in sect 1. > > | > | >From an Int-Area perspective, I feel some of the coverage of tunneling + > | multicast is lacking. For example, specifically, generation, treatment, > | defaults, throttling, processing of ICMP Error messages. The document does not > | include the acronym "ICMP", yet I believe it is important particularly for AMT. > | > | Further, combining this Int-Area perspective with some added operational > | aspects, manageability (from defaults to toolset) seem also highly lacking. > > Added Sect 9 to mention some operational considerations. > > > | > | I hope these are clear and useful. > | > | Thank you! > | > | Carlos. > | > | > | -- > | Mops mailing list -- mops@xxxxxxxx > | To unsubscribe send an email to mops-leave@xxxxxxxx > |
Attachment:
signature.asc
Description: Message signed with OpenPGP
-- last-call mailing list -- last-call@xxxxxxxx To unsubscribe send an email to last-call-leave@xxxxxxxx
