Re: [E-impact] Carbon-Intelligent Global Routing in Path-Aware Networks

[Rudolf van der Berg <[e-mail]>]

Hey Dom,

Thanks for looping me in. Particularly because I finished my presentation on the Energy use of the Internet only a minute ago, so I was presenting when I received it. I have attached the slides. The event will come online

RIPE 86 Student Event: Threats to the Internet as we know it and what the technical community (and you) can do to prevent them
Date: Tuesday, 9 May, 17:00 – 18:30 (UTC+2/CEST)
https://ripe86.ripe.net/programme/parallel-events/student-event/

Vesna asked me about Jevons as well 😊

Greetings,

Rudolf van der Berg
Partner
_____________________________
Stratix
'Villa Looverhoek' | Julianalaan 1
1213 AP Hilversum |the Netherlands
Tel:  [phone]
Mob:[phone]
E-mail: [e-mail]<mailto:[e-mail]>
URL: http://www.stratix.nl/

Van: Dom Robinson <[e-mail]>
Verzonden: dinsdag 9 mei 2023 18:11
Aan: Michael Welzl <[e-mail]>
CC: Vesna Manojlovic <[e-mail]>; Daniel Schien <[e-mail]>; [e-mail]; Paul Shabajee <[e-mail]>; Rudolf van der Berg <[e-mail]>
Onderwerp: Re: [E-impact] Carbon-Intelligent Global Routing in Path-Aware Networks

I’m adding Rudolf here - it was he who highlighted to me that modern fibre networks can be LESS power for many times more capacity - so a swift migration to such networks (and a ‘clean’ decommissioning / recycling of the legacy network)  effectively reduces the energy consumption.

He might have numbers he can share to validate this (to your point @Michael).

@Vesna: The Jevons paradox only comes in if the energy INCREASES as you increase capacity. If the energy REDUCES despite an increase in capacity then the Jevons paradox does not apply.

Also ANY ‘de-growth’ strategy needs to be clear about what it is we are growing or de-growing. In my mind the limited resource here is energy. That is where we should be focussing effort.

Incidentally @Michael: Wikipedia is wrong: 5g is only more efficient once you are using the majority of the network. In volumetric terms it demands 3 to 4 x MORE energy to deploy. If there is no traffic on the network then it is 3-4 time LESS efficient than not building it in the first place….
--
Dom Robinson
www.id3as.co.uk<http://www.id3as.co.uk/>
www.greeningofstreaming.org<http://www.greeningofstreaming.org/>
uk.linkedin.com/in/domrobinson<http://uk.linkedin.com/in/domrobinson>
Meet: https://calendly.com/id3as


On 9 May 2023, at 09:55, Michael Welzl <[e-mail]<mailto:[e-mail]>> wrote:
Hi !

I wanted to react to two things here - I’m cutting some other stuff:



In fact, counterintuitively there is a stronger argument that using “more data” commercially drives a swift migration to new technologies,

You say that as if it's a good thing ;-)

then all the older devices are either overheating, or have to be dumped, and have to be replaced with new devices... made from non-renewable materials, shipped across half-world, etc etc...

... which, of course, is out of the scope of what _we_ can do -- but, we can influence our own part of the process!

This raises an interesting question. Indeed, I always thought that “sustainability”, for us, should also mean to sustain the equipment itself (where we might have technically viable approaches: wouldn’t SDN / NFV allow for more re-usability?). But, there’s a trade-off, in that newer equipment is more energy efficient. So, what is better?  I also find it very hard to believe that we might better off if we throw away and re-produce more, faster. But … numbers, anyone?



which almost invariably, save energy once utilisation passes an equivalent point on the legacy network.  Newer IT is almost invariably more efficient.

I have to again refer to Jevons Paradox:
https://en.wikipedia.org/wiki/Jevons_paradox

"increases (in) the efficiency with which a resource is used ,
but the falling cost of use increases its demand,
increasing, rather than reducing, resource use."

For “efficiency”, we’re talking about energy efficiency here. I find the statement on this Wikipedia page regarding 5G extremely doubtful:
"5G is far more energy efficient than 4G. However, the lower cost will lead to more internet use, which in turn may result in an increase of energy consumption.[34]”

Internet use is growing because 5G is cheaper, because it costs less power?

Jevons paradox is interesting and certainly important to think about, but may not apply to everything everywhere. For networks, is there really any evidence that the usage growth is due to reduced energy consumption?

Cheers,
Michael





#catamongpigeons :)


(*)

Here's an interesting presentation, but I am not sure what the conclusion is -- I am curious what you read from it?

Energy optimisation in GÉANT
https://geant.app.box.com/s/budgt80vyjzlchxk16hs5v6b876r9iuf

X5 capacity for same power draw
but
Power growth ~10% p/a accounting for planned line cards



Regards,
Vesna

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