Re: [OPSAWG] New Version Notification for draft-opsawg-poweff-00.txt

["Marisol Palmero Amador (mpalmero)" <[e-mail]>]

Thanks Joe for your notes and valid comments.
Please find my answers inline

Marisol Palmero


From: Joe Clarke (jclarke) <jclarke=[e-mail]>
Date: Friday, 27 October 2023 at 20:53
To: Marisol Palmero Amador (mpalmero) <[e-mail]>, [e-mail] <[e-mail]>
Cc: Gonzalo Salgueiro (gsalguei) <[e-mail]>, Jan Lindblad (jlindbla) <[e-mail]>, Snezana Mitrovic (snmitrov) <[e-mail]>, [e-mail] <[e-mail]>, Per Andersson (perander) <[e-mail]>, Esther Roure Vila (erourevi) <[e-mail]>
Subject: Re: New Version Notification for draft-opsawg-poweff-00.txt
I had a brief read of POWEFF, and I have some questions/possible discussion points for your presentation (as a contributor).

You mention calculation in the draft, but I didn’t see this reflected in the data model.  Note that I’m not sure the data model needs to address calculations per se.  I think you could spell those out outside of YANG, with respect to the data objects.  I just expected to see some specifics or examples of using the modeled data to reflect a composite energy efficiency metric.


[Marisol]
Calculations will be part of what we consider part of the POWEFF-derived module
We need more discussions, and feedback from the group on it
Our next steps are considering linking into the draft-lindblad-tlm-philatelist proposal. This module should contribute to the data produced by the aggregation/processor, where we might need to consume from pointers and metadata, like ietf-interfaces model.

On the data model, I’m a bit confused why the network interface module is needed.  In your text about this you say that it might line up exactly to standardized modules, but it’s needed for those things that are not network devices.  What devices are not network devices that have network interfaces and don’t fit into the general, say, ietf-interfaces model?

[Marisol]
Our thought it is more in the direction on how to add into “useful work” and which attributes/“sensors” might be part of it. As part of the asset definition, we are not only looking into network devices, but we also consider servers, as part of compute, they also have defined interfaces, and it might be where YANG ietf-interface model doesn’t fit well. But we agree that pointers should address different types of assets.

Finally, why are all of these objects config true?  The way I read this, a vendor or device manufacturer would stream these objects based on how the system is built or how it’s operating.  I can’t see any reason why one would need to configure any of this.
[Marisol]
100%, we need to change this.


As chair, I’d love to see how this factors into the green networking metrics work that has been presented at opsawg before.  Some of the text touches on similar topics.  Perhaps POWEFF can be seen to take a practical approach to some of the concepts and thoughts in draft-cx-opsawg-green-metrics?

[Marisol]
We have scheduled a Side Meeting on Monday morning, and it will be great to review with the green-metrics authors, how we should move forward
Side meeting: Monday Nov 6, 8:45 - 9:30                Karlin 4 Sustainability Insights. Description: Gaps on Power Metrics Normalization


Many thanks,
Marisol



Joe

From: OPSAWG <[e-mail]> on behalf of Marisol Palmero Amador (mpalmero) <mpalmero=[e-mail]>
Date: Friday, October 27, 2023 at 13:07
To: [e-mail] <[e-mail]>
Cc: Gonzalo Salgueiro (gsalguei) <[e-mail]>, Jan Lindblad (jlindbla) <[e-mail]>, Snezana Mitrovic (snmitrov) <[e-mail]>, [e-mail] <[e-mail]>, Per Andersson (perander) <[e-mail]>, Esther Roure Vila (erourevi) <[e-mail]>, [e-mail] <[e-mail]>
Subject: [OPSAWG] FW: New Version Notification for draft-opsawg-poweff-00.txt
Dear OPSA WG,

Earlier this week, we posted a new draft that introduces a data model for power and energy related metrics :
https://datatracker.ietf.org/doc/draft-opsawg-poweff/



The focus is mainly on runtime information provided by power sensors, but also an extension to other related metrics and given attributes that will complement the representation of the energy consumed by the network device, implemented in hardware or software, as well as by specific network components.
This is a first-version approach where we see still challenges based on implementation.
Note: Some of those challenges are covered on Jan`s draft: draft-lindblad-tlm-philatelist<https://datatracker.ietf.org/doc/draft-lindblad-tlm-philatelist/>

Along with POWEFF draft, we’ve also updated the version of the Sustainability Insights draft:
https://datatracker.ietf.org/doc/draft-almprs-sustainability-insights/
where we introduced an updated architecture reference diagram, that provides a more structured view of the functional blocks that might be part of where those attributes and metrics might be produced, processed, visualized, etc. We also have reviewed and added Use Cases that such framework could drive.

We greatly appreciate your thoughts and comments.

Many thanks,
Marisol Palmero


From: [e-mail] <[e-mail]>
Date: Friday, 20 October 2023 at 17:45
To: Gonzalo Salgueiro (gsalguei) <[e-mail]>, Jan Lindblad (jlindbla) <[e-mail]>, Marisol Palmero Amador (mpalmero) <[e-mail]>, Snezana Mitrovic (snmitrov) <[e-mail]>
Subject: New Version Notification for draft-opsawg-poweff-00.txt
A new version of Internet-Draft draft-opsawg-poweff-00.txt has been
successfully submitted by Marisol Palmero and posted to the
IETF repository.

Name:     draft-opsawg-poweff
Revision: 00
Title:    Power and Energy Efficiency
Date:     2023-10-20
Group:    Individual Submission
Pages:    37
URL:      https://www.ietf.org/archive/id/draft-opsawg-poweff-00.txt
Status:   https://datatracker.ietf.org/doc/draft-opsawg-poweff/
HTMLized: https://datatracker.ietf.org/doc/html/draft-opsawg-poweff


Abstract:

   This document motivates and specifies a data model to report power
   and energy efficiency of an asset.  As highlighted during the IAB
   workshop on environmental impacts
   (https://datatracker.ietf.org/doc/html/draft-iab-ws-environmental-
   impacts-report-00), visibility is a very important first step
   (paraphrasing Peter Drucker's mantra of "You cannot improve what you
   don't measure").  During the workshop the need for standardized
   metrics was established, to avoid proprietary, double counting and
   even contradictory metrics across vendors.

   This Power and Energy Efficiency Telemetry Specification (POWEFF) is
   required to promote consistency across vendors and consumers, based
   on: 1.  The definition of datasets and attributes defining a common
   data model utilized by the standard calculation to yield power and
   energy efficiency value for any asset or network element.  2.  The
   standard calculations utilizing the specified datasets and attributes
   which will yield energy consumption and energy efficiency value for
   any asset or network element.

   The model provides information and data requirements for calculating
   the Power and Energy Efficiency for specific assets.  Assets can
   include hardware (physical or virtual), software, applications, or
   services.



The IETF Secretariat