Re: [E-impact] kWh/GB should be banned and the IETF should warn against its use

[Toerless Eckert <[e-mail]>]

Several years back i worked with a customer in selecting equipment for
video streaming. The metrics that had to be compared was on the box level

- GB/USD (how expensive the box is)
- kWh/GB (how much power draw -> how much electrical opex)
- GB/RU (how dense) - RU = Rack Unit

Where the GB of course where measured for the function in question (not
just IPv6 data packet forwarding).

It didn't help to be smaller than a competitor because your location
can afford only some certain amount of cooling per RU. Which differs based
on deployment.

Of course, the more expensive power becomes the more relevant the factor is
beyond the mere cost of cooling.

This type of calculation goes on everywhere where you start stuffing electrical
equipment into rooms and optimize cost. Whether its servers or network equipment.

Even when you compare networks, and they are generally very different in services/scope
or other factors, you can on a per-equipment basis compare the equipments energy
efficiencies. And many operators seem to be waking up and start doing this
renewed after the huge increase in energy costs due to the war in the Ukraine.

So, obviously, kWh/GB is a core factor required in any deployment of electrical 
equipment when you do have choices.  So an ask to eliminate the metric is just hilarious.

And when operators do not want to publish the numbers, it can simply be because
they know how bad they may loook on them.


Cheers
    Toerless

On Mon, May 15, 2023 at 09:26:44PM +0200, Rudolf van der Berg wrote:
> Alexander,
> 
> The reason I call for an absolute ban on the metric is because it is
> meaningless and misleading. It is not just that it shouldn't be the sole
> metric, it shouldn't be a metric at all. You are right to point out that
> Telefonica uses it. If you read this thread carefully, I have pointed out
> that the GSMA uses it too. The GSMA also shows that there can be a 10-20x
> difference between different network providers on this metric.Several
> telcos including Orange and KPN absolutely don't use it. Probably because
> their staff realised it is a nonsense metric. KPN does report energy use
> per network capacity. That already sounds more interesting.
> 
>  Furthermore it doesn't give actionable intelligence to anyone. What does
> it say if VodafoneZiggo in the Netherlands reports that its energy use went
> up from 160GWh to 180GWh while its energy use went down from 0.021kWh/GB to
> 0.013kWh/GB in its fixed Docsis based network after the Docsis 3.1 upgrade
> took place? I know the number of customers didn't increase dramatically,
> nor did its footprint change. Yet somehow with rising electricity use the
> metric shows a decline.
> https://rudolfvanderberg.medium.com/zet-alle-netwerken-uit-om-stroom-te-besparen-geen-netflix-en-geen-5g-d40f61eef0f0
> 
> As I explained there is no scientific basis for the metric. If you look at
> the work by professor Manner, you will see that using the metric to make
> prognosis about the Finnish mobile networks electricity use leads to
> estimates 3 times higher than the official figures.
> 
> the simple reality is that energy use and data traffic have no correlation
> in fixed and mobile broadband networks. It is like measuring the efficiency
> of cars based on wiper fluid use. Or traffic lights energy use based on the
> number of cars that drive by.
> 
> The IETF should do everything it can to develop RFCs that promote the
> efficient use of resources, it can't do that if it doesn't have valid
> metrics.
> 
> Sincerely,
> 
> Rudolf
> 
> Op ma 15 mei 2023 om 21:02 schreef Alexander Clemm <[e-mail]>:
> 
> > So, I am a little late to the thread.  Couple of comments:
> >
> > I do not agree with the premise of this thread that kWh/GB (or MWh/PB, as
> > used by Telefonica in their report) should be _banned_.  The only arguments
> > that I have heard seem to boil down to that relying on this metric as the
> > sole thing to guide us would be a bad idea.  I do agree with this; in fact,
> > any single metric can be gamed and will be misleading, and having a basket
> > of metrics will certainly paint a much more differentiated picture.
> > However, as one (not the only one) of many metrics I don't see it would be
> > harmful as the title of the thread suggests.  Telefonica uses this metric
> > in their report to track their progress over time - while comparing it
> > across different organizations with different contexts may result in
> > misleading apple-to-oranges comparisons, using it within the same
> > organization and presumably within the same context can be useful IMHO as
> > one way to indicate efficiency gains (and you can arguably see a measure of
> > their progress year-over-year (as indicated by the numbers they are
> > reporting pointing in the "right" direction) in their reports).   I think
> > it is fair for them to do so; my question to Rudolf would be, what is your
> > counter proposal as to what metrics service providers should use instead?
> >
> > The use of this particular metric is related to the overall energy
> > consumption of a service provider (the overall kWh/mWh/gWh) and putting
> > this into some kind of relation to an overall volume of service provided.
> > Of course, multiple questions: what is counted as part of the overall
> > energy consumption - e.g. just the power drawn by equipment, or the power
> > from their entire energy bill?  And, how is "volume" measured - why not
> > distance, etc?  Traffic volume in bytes also does not give a precise
> > indication of the user experience which may be better defined perhaps as
> > "hours of HD video streaming" or such (with actual traffic volume also
> > influenced by codecs, resolution etc), but it is still an indicator.  So,
> > yes, it is imperfect, and since it is a derived metric, it makes sense to
> > provide also the first-order metrics (in that case, overall energy
> > consumption & overall volume).  IMHO, clearly isn't a "one size fits all"
> > metric.  On a related note, metrics to assess the network-at-large need to
> > be arguably different metrics from those that are used to evaluate
> > individual transmission equipment.
> >
> > --- Alex
> > On 5/12/2023 12:05 PM, Rudolf van der Berg wrote:
> >
> > Hesham,
> >
> > Look at the title of the thread. The kWh/GB metic is nonsense. It doesn't
> > convey any comparable and actionable meaning. The numbers are incomparable
> > because traffic and energy use are in no way and have never been related.
> > How is it possible that with the exact same network and equipment the
> > energy use is stable when traffic goes up? Simple it is a useless
> > unscientific metric.
> >
> > Op vr 12 mei 2023 20:48 schreef Hesham ElBakoury <[e-mail]>:
> >
> >> The DT CR 2022 reports 91 kwh/terabyte. This is higher than 60
> >> kwh/terabyte. Is this because the 60kwh is for the mobile network and the
> >> remaining 31kwh is for the DT wired network?
> >>
> >> If DT uses .06 kwh/GB for mobile network,  that is better than the
> >> 0.1-0.6 kwh/GB range reported by GSMA. Does that mean that DT network
> >> energy efficiency is high or ... ?
> >>
> >> Thanks,
> >> Hesham
> >>
> >>
> >> On Fri, May 12, 2023, 8:22 AM Rudolf van der Berg <
> >> [e-mail]> wrote:
> >>
> >>> GSMA is mobile only..
> >>>
> >>> Op vr 12 mei 2023 17:18 schreef Hesham ElBakoury <[e-mail]>:
> >>>
> >>>> Thanks Rudolf!
> >>>> For DT the numbers are getting smaller every year. I would expect that
> >>>> the numbers in 2023 to be smaller than 2022.
> >>>>
> >>>> In a different email you mentioned that Deutsche Telekom reports that
> >>>> it uses 60kWh per Terabyte.  This is different from the numbers reported in
> >>>> 2022 CR report.
> >>>>
> >>>> You mentioned that of the participating operators, in this year GSMA
> >>>> report  "66% performed in the 0.1–0.6 kWh/GB range  meaning that most of
> >>>> the included networks used 0.1–0.6 kWh to transfer 1 GB of data. Values
> >>>> outside of this range can be explained by inefficient networks, consumer
> >>>> habits or low data traffic per connection". DT 2022 CR report numbers are
> >>>> smaller than 0.1 kwh/GB. How to explain this low value?
> >>>>
> >>>> Hesham
> >>>>
> >>>> On Fri, May 12, 2023, 7:32 AM Rudolf van der Berg <
> >>>> [e-mail]> wrote:
> >>>>
> >>>>> Here is annual report 2022
> >>>>> https://report.telekom.com/annual-report-2022/services/downloads.html
> >>>>>
> >>>>>
> >>>>> https://www.cr-report.telekom.com/2022/management-facts/environment/energy-consumption-efficiency#atn-19726-19728,atn-19726-19729,atn-19726-19730
> >>>>>
> >>>>>
> >>>>> Op vr 12 mei 2023 16:19 schreef Hesham ElBakoury <[e-mail]
> >>>>> >:
> >>>>>
> >>>>>> The DT CR-2021 report describes the value for kwh/terabyte [
> >>>>>> https://www.cr-report.telekom.com/2021/management-facts/environment/energy-consumption-efficiency#atn-19726-19729,atn-19726-19730,atn-19726-19733,atn-19726-19728]
> >>>>>>
> >>>>>>
> >>>>>> Note that the total energy consumption increases by 4% year over year
> >>>>>> while the kwh/terabit decreases year over year.
> >>>>>>
> >>>>>> I am not sure how the numbers in kwh/GB which are reported in this
> >>>>>> email thread are calculated?
> >>>>>>
> >>>>>> Hesham
> >>>>>>
> >>>>>>
> >>>>>>
> >>>>>> On Fri, May 12, 2023, 2:20 AM Romain Jacob <[e-mail]> wrote:
> >>>>>>
> >>>>>>>
> >>>>>>> So no DT is not overprovisioned and utilisation is actively
> >>>>>>> obstructed. Or do you have other data and sources that support your
> >>>>>>> assessment?
> >>>>>>>
> >>>>>>> Not at all. It was not an assessment, only a suspicion. I only
> >>>>>>> mentioned DT because I understood your previous email as suggesting that DT
> >>>>>>> was maybe not to blame for their poor kWh/GB numbers -- it seems I
> >>>>>>> understood wrong and DT is indeed to blame. I have no issues with that.
> >>>>>>>
> >>>>>>> --
> >>>>>>> Romain JACOB
> >>>>>>> Postdoctoral Researcher
> >>>>>>> ETH Zurich
> >>>>>>> Networked Systems Group (NSG)
> >>>>>>> Lead: Prof. Laurent Vanbever
> >>>>>>> www.romainjacob.net
> >>>>>>> @RJacobPartner <https://twitter.com/RJacobPartner>
> >>>>>>> @[e-mail] <https://discuss.systems/@jacobr>
> >>>>>>> Gloriastrasse 35, ETZ G81
> >>>>>>> 8092 Zurich
> >>>>>>> [phone]
> >>>>>>> --
> >>>>>>> E-impact mailing list
> >>>>>>> [e-mail]
> >>>>>>> https://www.ietf.org/mailman/listinfo/e-impact
> >>>>>>>
> >>>>>>
> > --
> > E-impact mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/e-impact
> >

> -- 
> E-impact mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/e-impact


-- 
---
[e-mail]