Re: [alto] Discussion on the future of ALTO WG

["Y. Richard Yang" <[e-mail]>]

Hi Sabine,

On Thu, Mar 23, 2023 at 8:58 AM Sabine Randriamasy (Nokia) <
[e-mail]> wrote:

> Hi Med, Qin and ALTO WG
>
>
>
> Thanks a lot for initiating this discussion and your options proposal.
>
>
> https://github.com/ietf-wg-alto/wg-materials/blob/main/FutureALTO/alto-direction-of-work.md
>
>
>
> I definitely prefer Proposal #3: Support ALTO extensions for the new
> industry needs
>
>
>
> Thanks a lot to Jordi's insights on this option, that I share. ALTO WG may
> also want to work on abstraction of compute metrics and exposure, in
> relation to other IETF WG that would look at these matters. As a previous
> example, draft-ietf-alto-performance-metrics-28 (in RFC Ed queue) was done
> in coordination with the IPPM WG.
>

Thank you so much for this positive, collaboration motivation suggestion.
Given the detail, it appears Option 3 can be "Support ALTO extensions for
the new industry *and other WG* needs", but ultimately it should be driven
by industry need and I am hence also fine with just the current title.

The important piece by Sabine which I want to single out is "To add another
motivation for option 3, it is to be noted that CATS in its current charter
is mandated to produce *Informational* documents while ALTO is aiming at
standardized compute & network infrastructure exposure." So this can be a
good collaboration opportunity as well, in addition to separation into the
networking domain and application domain in the computing integration use
domain.

Richard

>
>
> Kind regards,
>
> Sabine
>
>
>
> *From:* alto <[e-mail]> *On Behalf Of *
> [e-mail]
> *Sent:* Thursday, March 23, 2023 8:12 AM
> *To:* Jordi Ros Giralt <[e-mail]>; Qin Wu <[e-mail]
> >
> *Cc:* [e-mail]
> *Subject:* Re: [alto] Discussion on the future of ALTO WG
>
>
>
> Hi Jordi, all,
>
>
>
> Only some logistic comments, not reacting to any expressed views so far:
>
>
>
>    - We created a page at
>    https://github.com/ietf-wg-alto/wg-materials/blob/main/FutureALTO/alto-direction-of-work.md
>    to track the various proposals (yours is posted there), challenge them,
>    enrich them, add rebuttals, etc.
>    - For your logistic comment, we organized on purpose an interim
>    meeting to offload the IETF#116 agenda and let other I-Ds be presented and
>    discussed. We scheduled 4 other interims till end of May. We really need
>    some focus at this stage.
>
>
>
> Thanks.
>
>
>
> Cheers,
>
> Med
>
>
>
> *De :* Jordi Ros Giralt <[e-mail]>
> *Envoyé :* jeudi 23 mars 2023 00:13
> *À :* Qin Wu <[e-mail]>; BOUCADAIR Mohamed INNOV/NET <
> [e-mail]>
> *Cc :* [e-mail]
> *Objet :* Re: [alto] Discussion on the future of ALTO WG
>
>
>
> Hi Med, Qin,
>
>
>
> Here is my feedback to your analysis below.
>
>
>
> I would like to start with a note. The ALTO team has brought (and
> continues to bring) a lot of positive energy (development of RFCs,
> deployments of the standard at major carriers and new deployments that are
> in the making, running code via the development of the open source project
> OpenALTO, consistent participation on IETF hackathons usually with multiple
> parallel projects/demos, chairing important forums such as SIGCOMM NAI to
> incorporate feedback into the WG from the broad spectrum of industry and
> academic players, etc.), but it is also true that much of the (even larger)
> potential energy of the group has been locked for quite some time as the
> group has not been allowed to discuss the new critical topics that we want
> to bring from our industry needs. We've all being waiting for this moment,
> to be able to discuss the new topics and unlock yet another level of
> positive energy into the IETF; and so, it is at the minimum surprising that
> the only two options being proposed are either (1) recharter with just a
> focus of working on protocol maintenance or (2) close the WG and move our
> current work to other WGs or RGs.
>
>
>
> I have two broad comments, one on the proposed options and another one on
> the logistics to make a proper decision.
>
>
>
> On the proposed options:
>
> ------------------------------------
>
> I would like to suggest adding a 3rd proposal, which I believe is what
> much of us have been working for, for quite some time:
>
>
>
> # Proposal #3: Support ALTO extension for the new industry needs.
>
>
>
> ## Rationale:
>
>    - Many I-Ds have been proposed describing the importance of leveraging
>    ALTO key core architecture to enable the new industry needs, where close
>    cooperation between the application and the network is critical.
>    - Allowing these extensions would enable the group to grow and unlock
>    its true potential, also attracting other industry players that have been
>    writing ALTO I-Ds, but not fully joined us yet because their proposals were
>    tagged as being out of the scope for the current charter.
>    - Lots of positive energy and determination in the WG, as we
>    understand the potential positive impact (better application performance).
>    - The proposed work can't be done in other groups, and even if we
>    tried to do so, it would be improper from an architecture/engineering
>    standpoint. For instance, trying to move the exposure of compute
>    information for determining edge services to CATS is not viable since
>    "Exposure of network and compute conditions to applications is not in the
>    scope of CATS" [1]. ALTO is inherently/by definition very well
>    positioned here, since it's designed to expose such kind of
>    information to the application, that is key to the industry problems we are
>    working to resolve.
>    - There is a natural, coherent story for ALTO, which started from P2P
>    networks, then CDNs, and now it's moving into edge computing, where the
>    application requires more than ever to cooperate closely to meet stringent
>    throughput and delay requirements.
>    - There is a belief that the ALTO WG has been running for a very long
>    time, but this in general is not a good technical reason to base a
>    rechartering decision on. From the abovementioned trend standpoint,
>    keeping ALTO open to provide the IETF a platform for close
>    application-network integration appears more important than ever before.
>
> ## Proposed direction of work:
>
> ·Recharter the WG with a focus on ALTO to cover both maintenance and the
> new industry needs (where such needs are currently being discussed in the
> ALTO WG internal meetings and mailing list, see also my next comment on
> logistics).
>
> [1] CATS charter: https://datatracker.ietf.org/doc/charter-ietf-cats/
>
>
>
> On the logistics to make a proper decision:
>
> ---------------------------------------------------------
>
> This is of course a very important decision, so it's also important that
> we as a group provide the right discussion environment to make a proper
> decision. For instance, various members of the WG have been working on
> various I-Ds to enable a discussion of the proposed new charter items. Yet
> during IETF 116, the group is only given 20 minutes to discuss 5 different
> I-Ds that are proposed topics for the recharter. This is not sufficient
> time to enable a proper discussion on these important topics. Granted, the
> ALTO WG meets every week, and we can have further conversations offline,
> but the IETF Meetings are a great place to have these discussions in person
> and to open them up to people outside the group to collect feedback. I
> would encourage providing proper time while we are in Japan to discuss
> these topics and continue to discuss them thereafter via interim meetings.
> (During 116, getting 30 minutes would be better than 20, getting 40 minutes
> would be even better.)
>
>
>
> Thanks,
>
> Jordi
>
>
>
>
>
> _________________________________________________________________________________________________________________________
>
>
>
> Ce message et ses pieces jointes peuvent contenir des informations confidentielles ou privilegiees et ne doivent donc
>
> pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu ce message par erreur, veuillez le signaler
>
> a l'expediteur et le detruire ainsi que les pieces jointes. Les messages electroniques etant susceptibles d'alteration,
>
> Orange decline toute responsabilite si ce message a ete altere, deforme ou falsifie. Merci.
>
>
>
> This message and its attachments may contain confidential or privileged information that may be protected by law;
>
> they should not be distributed, used or copied without authorisation.
>
> If you have received this email in error, please notify the sender and delete this message and its attachments.
>
> As emails may be altered, Orange is not liable for messages that have been modified, changed or falsified.
>
> Thank you.
>
> _______________________________________________
> alto mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/alto
>