Re: [ippm] New Internet Draft: Congestion Signaling (CSIG)

Dear Abhiram and authors,

Thanks for posting this draft.

A few comments:
- The notion of using fixed-length end-to-end performance metrics over
data packets is interesting and under discussion in IPPM. Please have
a look at the following existing draft:
https://datatracker.ietf.org/doc/draft-mzbc-ippm-transit-measurement-option/
- The concept of transparently forwarding the CSIG-tag across routers
very roughly reminds me of the "uniform model" in RFC 2983, where QoS
fields are copied across tunnel encapsulating/decapsulating devices.
If we view the CSIG-tag as an extension of the IEEE 802.1Q PCP field,
then the current draft requires routers to forward the L2 QoS fields
when routing, which is a bit similar to the uniform model.
- Having said that, I would suggest you reconsider the comments from
previous reviewers about considering the CSIG-tag in the IP layer. You
may try to define an alternative encapsulation of the CSIG-tag as an
IPv6 hop-by-hop option, and see whether you lose something by using
this approach.
- Is it necessary to require that the CSIG-tag is the innermost tag? Why?
- The MACsec security-tag needs to be discussed further. Generally
speaking, everything after the security-tag is encrypted and/or
integrity-protected, so specifically the CSIG-tag can't be modified by
802.1 bridges if incorporated after the security-tag. You may consider
placing the CSIG-tag before the security-tag.
- Please note that coexistence with FGL tags (RFC 7172) and E-TAGs
(IEEE 802.1BR) should also be considered.
- Regarding CSIG Reflection, it would be useful to discuss how it
would work with RDMA.

Cheers,
Tal.

On Sat, Sep 9, 2023 at 5:43 AM Abhiram Ravi
<abhiramr=40google.com@dmarc.ietf.org> wrote:
>
> Hi IPPM folks,
>
> I am pleased to announce the publication of a new internet draft, Congestion Signaling (CSIG): https://datatracker.ietf.org/doc/draft-ravi-ippm-csig/
>
> CSIG is a new end-to-end packet header mechanism for in-band signaling that is simple, efficient, deployable, and grounded in concrete use cases of congestion control, traffic management, and network debuggability. We believe that CSIG is an important new protocol that builds on top of existing in-band network telemetry protocols.
>
> We encourage you to read the CSIG draft and provide your feedback and comments. We have also cc'd the TSVWG, CCWG, and ICCRG mailing lists, as we believe that this work may be of interest to their members as well.
>
> Thank you for your time and consideration.
>
> Sincerely,
> Abhiram Ravi
> On behalf of the CSIG authors
> _______________________________________________
> ippm mailing list
> ippm@ietf.org
> https://www.ietf.org/mailman/listinfo/ippm