On Jun 16, 2022, at 9:22 AM, Sebastian Moeller <moeller0@gmx.de> wrote:

HI MIchael,

On Jun 16, 2022, at 07:48, Michael Welzl via Bloat <bloat@lists.bufferbloat.net> wrote:

… but I’m excited about slide 15 !

"Heat, CO2, and radioactive waste are becoming measurable by-products of TCP inefficiency
Fix TCP => Save the World!”

People don’t yet focus enough on this -

The slide deck however omits presenting the evidence for that hypothesis...

and it’s not only relevant in a data center context. See also:

Michael Welzl: "Not a Trade-Off: On the Wi-Fi Energy Efficiency of Effective Internet Congestion Control", IEEE/IFIP WONS 2022, virtual, 30 March - 1 April 2022.
Preprint: https://folk.universitetetioslo.no/michawe/research/publications/wons2022_authors_version.pdf

A couple of IETFs ago, I did sign up to present something along these lines at an ICCRG meeting, but I asked late and got squeezed into an “if time permits” slot; some day in the future I’ll ask for a “proper slot” to elaborate on this.

Curious, should such an observation not look at the aggregate energy consumption of senders and receivers? And maybe also report the total energy consumption of those devices for reference to allow an estimate of how much "radioactive waste" can be easily avoided?
Mind you, I am not arguing that saving energy (even small amounts) is not desirable, I just want to see how that stacks up with the total energy consumed by those systems.