[irtf-discuss] Re: [E-impact] [Green] Re: Network Equipment Energy Efficiency Metric

[Toerless Eckert <[e-mail]>]

On Tue, Nov 12, 2024 at 12:46:48PM -0800, Tony Li wrote:
> Hi Toerless,
> 
> 
> > On Tue, Nov 12, 2024 at 10:40:11AM -0800, Tony Li wrote:
> >> [-green, as this is inappropriate]
> > 
> > ? You don't think green WG members would find the explanation you provided
> > useful to the extend that they may wonder if/what of it they want to see
> > modelled in their work ?
> 
> 
> No, since this is all internal behavior and does not have any bearing on anything that we should be standardizing.

I think one wants to understand and model the aspects that change energy consumption
so that whoever observes the IETF model can know the root cause for the current
enerrgy consumption.

> >> Data sheet power usage for networking gear is typically peak, worst-case power utilization, as this is necessary for provisioning purposes.
> > 
> > I've also not always found this being broken apart into the modular components, so
> > you may only know what the maximum config (e.g.: all-fastest-linecards) would
> > consume. I guess/hope green-WG would have that info though.
> 
> 
> It’s wholly out of scope for IETF to collect manufacturer data.  It is up to the manufacturers to provide that data, hopefully through GREEN specified models.

I think we're in violent agreement, but you may have misunderstood me:
If the IETF model appropriately breaks
down the energy consumption and manufacturer implement the model, then
we may get the info that was previously inconsistently provided by manufacturers.

Cheers
    Toerless
> 
> 
> T