Re: [E-impact] [OPSAWG] side meeting #119: Power Metrics: concrete usage example

[Carlos Pignataro <[e-mail]>]

Hi, Rob,

Thanks again for the thoughtful responses -- please also see inline.

[Hi, Suresh, please find one small parenthetical note for you inline as
well]

On Tue, Mar 26, 2024 at 6:28 AM Rob Wilton (rwilton) <[e-mail]>
wrote:

> Hi Carlos,
>
>
>
> Thanks for the comments.  I’ve provided some comments (RW) inline …
>
>
>
> *From: *Carlos Pignataro <[e-mail]>
> *Date: *Monday, 25 March 2024 at 21:09
> *To: *Rob Wilton (rwilton) <[e-mail]>
> *Cc: *Marisol Palmero Amador (mpalmero) <mpalmero=
> [e-mail]>, Ops Area WG <[e-mail]>, E-Impact IETF
> <[e-mail]>, [e-mail] <[e-mail]>,
> Alexander Clemm <[e-mail]>, Alberto Rodriguez-Natal (natal) <
> [e-mail]>, Ronald Bonica <[e-mail]>, Mahesh
> Jethanandani <[e-mail]>, Ali Rezaki (Nokia) <
> [e-mail]>, Suresh Krishnan (sureshk) <[e-mail]>,
> Jari Arkko <[e-mail]>
> *Subject: *Re: [OPSAWG] side meeting #119: Power Metrics: concrete usage
> example
>
> Hi, Rob,
>
>
>
> Thanks for the comprehensive email, and for your desire to support the
> industry towards improved energy efficiency!
>
>
>
> RW: Great!  I think that at least our broader goals are aligned here,
> although you seem to disagree on the particular path that I’m pushing for.
> I’m hoping that we can manage to get alignment, working towards the common
> good.
>
>
>

Rather than disagreeing, I must say I am confused.

I imagine you are subscribed to e-impact as well, since some feedback
towards the lack of value of WG formation from participants was sent there
but not Cc'ed to Opsawg.


>
>
> My first reaction is that this direction seems counter to and in conflict
> with the conclusion and decisions from the IAB Program eimpact “interim”
> from just a month before:
>
> ·         See Chair Slides
> <https://datatracker.ietf.org/meeting/interim-2024-eimpact-02/materials/slides-interim-2024-eimpact-02-sessa-chair-slides-01>,
> that codified: "*Metrics – Push through the WGs*” (etc. etc.)
>
> ·         See Minutes
> <https://datatracker.ietf.org/meeting/interim-2024-eimpact-02/materials/minutes-interim-2024-eimpact-02-202402161500-00>,
> that captured: "*Suresh agreed and mentioned that the reason for having
> the drafts here is that people to get higher level view since all working
> groups need to have a sustainability angle*"
>
> RW: Sorry, I had a conflict and couldn’t attend the e-impact interim.  My
> previous understanding was based on when this was discussed between the
> IESG and IAB retreat last summer was that the IAB program was for more
> future looking working, and incubating ideas that were not yet ready to be
> standardized but any actual work would happen in IETF WGs.
>

No worries, that's why I included excepts from the proceedings.


>
>
> A second thought is that, while on the surface getting a couple of
> document with ‘green metrics’ is useful and might seem net-positive,
> knee-jerk reacting on tactics misaligned with strategy can further fragment
> the Eimpact work (which already can be characterized as ‘having a hard time
> finding itself’ with work from 2022 and no output).
>
>
>
> RW: Sorry, but I don’t really follow.  Why would standardizing metrics and
> power controls now impact the overall strategy?  This is perhaps where I
> see things quite differently.  I see this as a simple split between what we
> can standardize now, relatively quickly, starting to reap the benefits now
> vs spending a long time discussing what we plan to do before taking any
> action.
>

Sorry if I was not clear -- my point is that a top-down approach on
analyzing what focus areas / work areas have the largest e-impact seems
valuable yet bypassed. Metrics was a key topic from the beginning, and
clearly something to solve. There does not seem to be full alignment on
e-impact on what metrics are more important.


>
>
>
>
> There are clear risks like (1) believing that metrics/models are the
> ultimate goal of “eimpact/green’ work, while (as mentioned on eimpact)
> there’s no analysis of the most useful focus area, and (2) forgetting what
> Suresh wrote that many WGs need ‘green’, and this would separate work in a
> corner, as opposed to embedding and integrating it.
>
>
>
> RW: I don’t believe that metrics/models are the ultimate goal at all, but
> they do seem like a useful first step.  Further, the purpose of this
> proposed WG isn’t really to create new work, but to better corral the
> existing work that folks are already trying to get started within the IETF
> now, and as I see it, struggling to get traction.
>
>
>

That can be said about every piece of e-impact work. I appreciate and value
you take real action in pushing forward a subset of that which falls under
Opsawg. Really. Other ADs in other areas are not doing the same, which
would add to the huge imbalance that e-impact (iab program) already has.


>
>
> A third thought is that we had asked for a (broader and more e-impactful)
> WG a year ago, and that was shot down in favor of this IAB Program :-|
>
>
>
> RW: When I was an AD, both in the previous side meetings and in the
> IESG/IAB retreat I was also a vocal proponent for creating a WG, and yes,
> the agreement at that time (9-10 months ago) that we should start with the
> IAB e-Impact program, and that the work could proceed in existing WGs.
>

This matches my recollection perfectly.

[parenthetical: Suresh Krishnan, here you can see what I reference of WG vs
IAB program, and how the IAB program was the decision not favoring the WG]


> RW: However, I’ve since seen “green” related drafts being presented in
> OPSAWG, within IVY and it was on the agenda for NETMOD.  I.e., it looks to
> me like there is work ready to progress now but looking for a good home.
> The issue here is that this work isn’t obviously and clearly in charter for
> any of these WGs except maybe OPSAWG.  IVY is meant to be specifically
> focussed on a base inventory YANG model and should concentrating on that
> task until it is complete. Alas, one of the downsides of OPSAWG is that
> it’s made up of different groups of individuals working on their own topics
> and lacks the cohesiveness and collective direction that a dedicate WG
> could provide.
>

I clearly see this issue, and I agree that the current state and operating
model is quite suboptimal.

If we as a broader community see that there are glaring gaps in what the
IAB e-impact program is able to achieve based on its framing, why not
tackle the root cause?

Do you think we could end up with a green-metrics wg, a green-routing wg, a
green-security wg?

And is there a risk that the experts on "green" might not be in green-bof
or every green-wg?

And to be explicit (and hopefully not repetitive), this specific topic of
e-impact really pushes the model given how interdisciplinary it is.


>
>
>
>
> Fourth, ‘green-bof’ is very very broad, while I understood your desired
> scope to be narrow. This would eclipse eimpact as the shinny new ball, and
> would potentially confuse people on where to participate (outside the lucky
> ones that attended a side meeting)
>
>
>
> RW: green-bof is just a list name.  I don’t really care what the BOF or WG
> is called.  My intention is that the scope of the WG that I’m trying to
> create is that it will be narrow to the work items that are achievable in
> the short term.  Regarding participation, for me, I would suggest that
> interested folks may wish to participate in both: E-impact for overall
> strategy and longer-term considerations of how we should evolve our
> protocols.  The new WG for short term focused work on green related work
> that is understood well enough and that we can get standardized now, in the
> short term.
>

My feeling, FWIW, is that there is a high-risk of losing synchronicity
between these two sets, long/short term, iab-program/green-bof.


>
>
> Fifth, and Lastly — frankly I was debating with myself whether to mention
> this privately or not, but since you brought it up and opened the topic —
> another issue. Backdrop: BOF and WG-forming suggestions were sent to
> /dev/null favoring the IAB Program as the solution. What follows is a set
> of factual observations and no judgement or intentionality attached to
> them. But there’s (1) cisco proponents and cisco side-meeting organizer
> despite the eimpact interim, (2) with a Cisco-only I-D [1], (3) a Cisco AD
> meeting with (4) a Cisco IAB Member, in the (5) historically least attended
> meeting, and change direction 180 degrees… Again, no extrapolation or
> conclusion, but even from an appearance or optics perspectives.
>
>
>
> RW: Right, my drive for encouraging this work is not as Cisco employee,
> but was as an AD, and as someone who feels that environment impact is a
> significant issue to humanity and I would rather the IETF takes steps
> towards standardizing something now in the short term (noting that it can
> always evolve over time) than to spend 3 years having discussions before
> taking concrete action.
>
>
>
> Because of the timing of the side meeting, the discussion didn’t really
> end up being about particular drafts at all, but instead focused on what
> drafts currently exist in this space.  My brief, very rough notes that I
> captured were:
>
> Alex:
>
> Green Networking Metrics OPSAWG
>
> Data models could be derived
>
> NMRG problem statement
>
> Sustainability considerations
>
>
>
> Luis:
>
> APi for getting energy data from routers, SD WAN focus
>
>
>
> Tony Li:
>
> Traffic engineering, optimize across the network, greatest benefit is
> turning off parts of network device rather than just reducing traffic.
> YANG module to disable links across the system.
>
>
>
> Qin: + Carlos (BMWG draft currently):
>
> Controlller level model, reporting and configuration, benchmark and test
> invididual devices. Reference work from other SDOs
>
>
>
> Marisol/Jan:
>
> Sustainability insights doc
>
> Telemetry philatelist. (collecting telemetry data in general + green
> metrics + device level control
>
> Power - Data model for devices, YANG based, or other protocols
>
> Telemetry to TSDB mapping doc
>
>
>
> RW: I.e., it seems to be me that there are multiple folk from different
> companies who are all interested in directly working on this, and this is
> why I suggested within that meeting that I thought that it would be helpful
> to target a WG forming BOF, with a narrow charter to focus on the items
> achievable in the short term.  My goal is to help connect these folks
> together and help give them the tools to help succeed.  For me, the
> benefits of a dedicate WG are: (i) Interested parties know where to go, and
> where to target their work, where to have discussions (ii) Guaranteed
> dedicated agenda time rather than competing with all other topics happening
> within these WGs.  (iii) an easy ability to hold virtual-interims to
> progress this work more quickly.
>
>
>
> Yes, I continue contributing in the industry and field to this topic, and
> I would cautious you consider a bigger picture to see what approach(es)
> actually help.
>
>
>
> RW: Sorry, I still don’t follow.  Reporting power metrics off devices (to
> management agents can get better current information about what the actual
> power usage is), and the ability to selective disable ports, forwarding
> ASICs, linecards, etc … seems like an obvious first step.  What is the
> bigger picture that I’m missing here?
>
>
>

Sorry I was not clear. There is environmental-expertise that needs to be
embedded in what specific metrics are useful, impactful, etc. It seems to
me (and welcome feedback), that while e-impact converged on the need of
metrics, there's less alignment as to what specifically some of those need
to be (for them to be really useful).


>
>
> I hope and trust these are useful and clear,
>
>
>
> RW: Sure.  I still believe that this is significant interest and energy to
> target a WG forming BOF for IETF 120.  Perhaps, initially, you would be
> willing to participate and help in that effort to scope what such a WG
> could look like?  If, after we get into the details, you still believe that
> it is the wrong path, or we can’t get consensus on what work we want to do
> then you can of course drop out at any time, or indeed speak at the BOF (if
> it gets scheduled) to indicate why you think that this isn’t the best path
> forward.  Does that seem like a reasonable/pragmatic way forward?
>
>
>

Thanks, Rob.

I am still v concerned that this work (announced on the e-impact mailing
list 1 day before IETF119 started (!!!), and happening at a time convenient
to who was already in Brisbane) risks to be completely separated from and
without any check-and-balance mechanisms to keep lockstep and synchronicity
between what you described as green long vs. short term considerations
(paraphrasing, e-impact and green-bof, respectively.).

Carlos.

Regards,
> Rob
>
>
>
>
>
> Best,
>
>
>
> Carlos.
>
>
>
> [1] I did not see a response to this:
>
>
>
> RW: I’m not familiar with this draft/work, but I only see Cisco authors
> listed currently.  But my understanding is that others have proposed
> similar drafts leveraging similar ideas, e.g.,
> draft-cwbgp-ivy-energy-saving-management-01, that you are an author on, and
> which I also haven’t read …. However, it is also worth noting that
> draft-cwbgp-ivy-energy-saving-management-01 is targeted at the IVY WG,
> which I believe that is outside the scope of the current IVY WG charter, at
> least until the core inventory YANG model is completed and the WG
> recharters.
>
>
>
>
>
>
>
> *Poweff authors,*
>
>
>
> Is Poweff still a Cisco-only effort, as recorded in
> https://youtu.be/m4vpThE5K9c?feature=shared&t=3534? Verbatim youtube
> transcript:
>
> *Many of the um products uh that we have uh mainly in **Cisco** right we
> are still looking into multivendor and this will be really good for um the
> participants to um provide feedback how this H um standardization of the
> data model might impact in your network equipment but um*
>
>
>
> Thanks!
>
>
>
> Carlos.
>
>
>
>
>
>
>
>
>
>
>
> On Mar 25, 2024, at 10:48 AM, Rob Wilton (rwilton) <[e-mail]>
> wrote:
>
>
>
> Hi Carlos,
>
>
>
> During IETF 119, I had a couple of discussions with Suresh and Mahesh
> regarding how we actual get some of the short term “green” related work
> happening in IETF to get critical mass and cross review and get published
> in the short term.  This seemed to somewhat culminate during the Power
> Metrics side meeting where it is clear that:
>
> ·         Various folks, representing different organizations, have
> various drafts related to Green networking.
>
> ·         Currently these drafts are spread out to different working
> groups, have various amounts of overlap, and it is unclear that they
> currently have a good homes and sufficient traction in IETF to progress
> effectively.
>
> ·         There was support in the meeting to target a WG forming BOF for
> IETF 120 to create a new WG with a limited targeted charter.
>
>
>
> Hence the proposal from Suresh and I was to try and help coordinate for a
> WG forming BOF for IETF 120 scoped specifically to work on items that are
> understood and achievable in the short term.  E.g., roughly, I currently
> think of this work scope as being: e.g., energy related terminology and
> definitions (that should try and leverage and reference existing
> definitions from existing published sources), reporting energy and
> sustainability at the device and network layer via operational YANG models,
> and to facilitate configuration or YANG RPCs to influence and optimise
> power usage on network devices.  Longer term energy efficiency and Green
> networking goals are intended to be out of scope for the proposed WG’s
> initial charter, and should continue to be discussed as part of the
> E-Impact IAB program.  The exact scope of the charter would be worked out
> between the interested parties in the coming weeks.
>
> I’m happy to try and help this work gain traction within the IETF.  I
> appreciate that several of the proponents for this work are also from
> Cisco, but I have no vested interest other than trying to help the industry
> take small steps that may help improve energy efficiency in networks (e.g.,
> reporting power usage, and as Tony suggests by selectively powering off
> ports or linecards) to try and help mitigate some of the impacts of the
> Internet on climate change.
>
> To that end the proposed next steps from that side meeting were:
>
> 1.      For me to request the creation of new open “green-bof” mailing
> list from Mahesh (hopefully should be done over the next few days).
>
> 2.      I asked for, and received, permission to subscribe those who
> attended the side meeting, but once created, I also intended to circulate
> the existence of the mailing list to e-impact, and other places where
> related discussions have been taking place, so that others can join.
>
> 3.      To create a github location where we can reference drafts and
> collecting work on a BOF proposal and draft charter for the WG (which as I
> stated above, should be narrowly scoped to only the work that is well
> understood and achievable in the short term).  If I can get this under the
> IETF github space, great, otherwise I can host a personal github.  I’m
> already checking with Mahesh on the feasibility of the github location
> being IETF hosted.
>
> 4.      Once the mailing list is up and running, the next step is to
> arrange a few virtual meetings to try and gain consensus on the proposed
> initial scope of the WG, and to start reviewing and pulling together the
> BOF proposal, and charter text.
>
> 5.      To submit a BOF request for IETF 120.  The key dates being:
>
> a.      Warn the IESG and Secretariat that we are hoping for a BOF by 22nd
>  April (note Mahesh is already aware and this has already been informally
> flagged to the IESG)
>
> b.     Get the initial BOF submission in before 5th May
>
> c.      Refine the BOF proposal based on feedback received, and update by
> 7th June
>
> d.     14th June, we hear back whether the BOF has been approved for IETF
> 120
>
> e.      Continue prepping slides, etc, for the BOF, running up to early
> July
>
> 6.      In my experience, despite it being 4 months between IETF
> meetings, the time invariably disappears quickly, so I think that we need
> to frontload the BOF preparation effort to achieve consensus at IETF 120
> for creating a working group.
>
>
>
> Anyone else in the side meeting, please feel free to add anything that I
> have missed, or correct me, if I have misrepresented anything.
>
>
>
> Carlos, hopefully you are also interested in participating in these
> efforts.  If you have any feedback on the planned approach I would be glad
> to hear it.
>
>
> Regards,
> Rob
>
>
>
>
>
> *From: *OPSAWG <[e-mail]> on behalf of Carlos Pignataro <
> [e-mail]>
> *Date: *Monday, 25 March 2024 at 12:01
> *To: *Marisol Palmero Amador (mpalmero) <
> mpalmero=[e-mail]>
> *Cc: *[e-mail] <[e-mail]>, [e-mail] <
> [e-mail]>, [e-mail] <[e-mail]>,
> Alexander Clemm <[e-mail]>, Alberto Rodriguez-Natal (natal) <
> [e-mail]>, Ron Bonica <[e-mail]>, Mahesh Jethanandani <
> [e-mail]>, Ali Rezaki (Nokia) <[e-mail]>,
> Suresh Krishnan (sureshk) <[e-mail]>, Jari Arkko <
> [e-mail]>
> *Subject: *Re: [OPSAWG] side meeting #119: Power Metrics: concrete usage
> example
>
> +Jari
>
>
>
> Hello,
>
>
>
> *Suresh, Jari,*
>
>
>
> I'm confused by this bullet point:
>
> *•              next steps? E.g. WG coordination/status, form a WG Design
> Team, call for a BOF?*
>
>
>
> Could you please clarify?
>
>
>
> I understood there's no WG (and hence no WG coordination nor status), in
> favor of the IAB Program. There cannot be a WG Design Team without a WG. I
> cannot find "design team" or 'BOF" (WG forming or not?) in the minutes of
> eimpact meetings <https://datatracker.ietf.org/program/eimpact/meetings/>,
> maybe I missed it.
>
>
>
> Is this an effort parallel to eimpact or a shadow meeting?
>
>
>
> *Poweff authors,*
>
>
>
> Is Poweff still a Cisco-only effort, as recorded in
> https://youtu.be/m4vpThE5K9c?feature=shared&t=3534? Verbatim youtube
> transcript:
>
> *Many of the um products uh that we have uh mainly in **Cisco** right we
> are still looking into multivendor and this will be really good for um the
> participants to um provide feedback how this H um standardization of the
> data model might impact in your network equipment but um*
>
>
>
> Thanks!
>
>
>
> Carlos.
>
>
>
> On Fri, Mar 15, 2024 at 1:30 PM Marisol Palmero Amador (mpalmero)
> <mpalmero=[e-mail]> wrote:
>
> Dear all,
>
>
>
> We have booked a side meeting in Brisbone,  IETF #119
>
> *Thursday 9:00 am local time*.
>
> *Headline*: Power Metrics: concrete usage example
>
>
>
>
>
> Please see the *agenda* that we are proposing:
>
>
>
> •              Overview of ongoing sustainability work in IETF (everyone
> contributes)
>
> •              Brief presentation of sustainability insights/poweff
> updates, incl. look at a more concrete example
>
> •              Any other short updates?
>
> •              next steps? E.g. WG coordination/status, form a WG Design
> Team, call for a BOF?
>
>
>
>
>
> As we would like to leave time to discuss and review **next steps**, for
> the overview we propose not more than 20 min.
>
> As authors from specific drafts, please let me know which draft(s) you
> would like to review, we would like to make sure that we could fit them
> into the 20 min
>
>
>
> Safe travels, and have a nice weekend
>
>
>
> Marisol Palmero, on behalf of the authors of sustainability insights&
> poweff drafts
>
>
>
> _______________________________________________
> OPSAWG mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/opsawg
>
>
>