20
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
this post was submitted on 23 Jun 2023
20 points (95.5% liked)
Solarpunk
5393 readers
147 users here now
The space to discuss Solarpunk itself and Solarpunk related stuff that doesn't fit elsewhere.
Join our chat: Movim or XMPP client.
founded 2 years ago
MODERATORS
I was recently at a conference for AWS (Amazon Web Services, AKA the cloud provider for a HUGE chunk of the internet), and part of the keynote claimed that it was greener to run in the cloud because... uh... well, they didn't exactly say. Don't get me wrong, I could see how it would be easier to make all AWS data centers compliant with using green energy than it would be to convince every random financial institution that their on-premises servers need to be green, but quite frankly it's Amazon and I don't trust that they're telling the truth about themselves and not just greenwashing.
Quite frankly, for things like lemmy instances, I think we could totally achieve a totally solar powered setup easily... but not easily at scale or reliably.
I've thought about how cool it would be to have a server room linked up with a solar array and batteries, and basically only have the servers up when there's enough energy to power them. In theory, it sounds fun to have a static splash page that shows when the servers are down that explains why they're down, as a way to make people think about how energy-expensive servers are. In practice, it sounds like a nightmare for a ton of reasons to have an intentionally flaky server. But it sounds like this is already a Thing with Low-tech magazine, which is neat!
But that's not to say we couldn't build and self-host a reliable and sustainable server room. Just that I don't know the numbers on what a server room actually pulls energy wise and how much energy generation we'd need.
What is often omitted is that large centralized data-centers need a lot of cooling. Due to efficiency improvements this has somewhat improved lately, but it used to be up to 60% of the total electricity used.
Smaller decentralized servers don't need nearly as much of it as they can easily dissipate heat to their cooler surrounding even if they use older less efficient equipment.
Thus up-cycling older server hardware in decentralized locations can save a lot of energy if you consider the entire life cycle of the equipment.
Also omitted - the amount of speculative buying for planned capacity that never actually happens. I worked for one of the big tech companies for several years, and more specialized hardware especially (ML accelerators) were spun up with the notion of "we don't know who will need these, but we don't want to not have them if they're needed". Cue massive amounts of expensive hardware sitting plugged in and idle for months as dev teams scramble to adopt their stuff to new hardware that has just enough difference in behavior and requirements as to make it hard to migrate over.
Also also, there's a bunch of "when in doubt, throw it out" - automated systems detecting hardware failure that automatically decommission it after a couple strikes. False positive signals were common, so a lot gets thrown out despite being perfectly fine.