"While we at Microsoft have worked hard to be 'carbon neutral' since 2012, our recent work has led us to conclude that this is an area where we’re far better served by humility than pride," wrote Smith in the blog. "And we believe this is true not only for ourselves, but for every business and organization on the planet."
Such a beautiful way of leading -- leading with admission of failure
90% of the CO2 in a data center is caused by its infrastructure and IT architecture. Only 10% is operational (electricity, etc.). But we only seem to talk about the electricity usage. To understand digital CO2, analyze the materials used to make the devices.
staggering
We’re not prepared for the end of Moore’s Law
diminishing benefit of upgrading to efficient hardware
The Internet Eats Up Less Energy Than You Might Think
cloud energy usage -- network externalities not a big deal?
Why green cloud optimization is profitable for you and the planet
Decent but basic introduction to cloud greening
Cloud Carbon Footprint
useful calculator maybe
Energy-Efficient Peer-to-Peer Computation Offloading Based on Non-Causal CPU-State Information
Could carry some meaningful insights for Holo hosting
Energy Efficiency Dilemma: P2P-cloud vs. Datacenter
blip bloop
Energy Efficiency of P2P and D
blop blop
Global Energy Review 2020
Lotsa big data in here, but I found the World In Data sources much more useful.
Energy consumption by source, World
People really love substitution method, though, which shows the proportion of 'useful' energy. Dunno how I feel about this re: telling the climate change story, because inefficiencies definitely ought to be accounted for.
Global direct primary energy consumption
Direct method is more representative of carbon emissions, I feel
Duck curve
problem with solar energy generation
Dispatchable generation
coordination problem for renewable-heavy systems; gets even more complicated when generators are comunity run
HydropowerNuclearOther renewables
Not lookin' so great there
Where do greenhouse gases come from?
For Canada, they muddy petroleum consumption by mixing electricity and heat together. Frustrating!
Another observation is that much of that 28% on transportation could be electrified -- which is a net loss if we stick with petroleum and a net gain (maybe) if we move to green
Carbon dioxide emissions byend-use sectors, 20201,6289493192273363578504
breakdown of sectors -- 31.7% of GHG over here
Click to enlarge
Electricity accounts for 32% of US GHG emissions
3M Fluorinert FC-72
GWP of 9300 https://en.wikipedia.org/wiki/Perfluorohexane
results in the use of much less energy.
Great point. I wonder if recovery is thorough enough to prevent escape into the atmosphere.
clean, environmentally friendly
haha nope
Drought-stricken communities push back against data centers
Holy cow -- not just electricity but water too! Starting to be convinced by Gerry McGovern's argument that the only right thing to do is to reduce our digital consumption overall.
immersion cooling, where servers are submerged in a liquid that boils at a lower temperature than water, taking the heat with it.
One solution uses Fluorinert FC-72, which has a GWP of 9000 when it's not sealed or recovered. https://www.gigabyte.com/Solutions/Cooling/immersion-cooling
There are better liquids which have a GWP of only 1.
Less water and electricity, more GWP.
recycled wastewater
This, on the other hand, is very cool.
developed underwater or partially submerged data centers
Are rising ocean temperatures a concern?
Onshore Wind Turbines Electricity 47.21 147.72 Utility-Scale Solar Photovoltaics Electricity 42.32 119.13
Fairly important in the aggregate scale; that's surprising -- I do wonder about the immediate CO2 cost of building all those new generators over the course of one single decade (capital vs marginal CO2 cost).
Based on recent projections, electrification of energy services, combined with the increased use of renewables and improvements in energy efficiency, could deliver a whopping 90% of the reduction in energy emissions that we need to achieve to meet global climate targets.
whoa, if true, that'd be a big deal.
It is estimated that for every dollar spent on energy efficiency we avoid spending more than $2 on energy supply. Yet globally, progress on energy efficiency has been slow – on average, 2% annually.
Reduction is critical, super effective, and not being addressed adequately -- room for growth!