91–100 of 172 posts
Earlier quoted context omitted.
More or less than burning fossil fuels to keep the data center cool at non-polar latitudes?
Less than or equal to the same amount of energy would be dissipated, but it would be closer to the actual ice.
I'm all for exploring new approaches, however: a) Salt water is highly corrosive. Wouldn't maintenance costs be high? b) Isn't marine biology highly sensitive to heat pollution? For high latency services like Amazon Glacier, wouldn't it make sense to host in a place like Iceland? Really cheap hydrothermic/clean power. Highly educated local talent pool, and relatively consistent cool temperatures. If you're maintainin…
> b) Isn't marine biology highly sensitive to heat pollution? Impact on the marine environment is briefly mentioned in the last couple of paragraphs of the article, although they're a little dismissive and don't provide any evidence supporting their claim of negligible impact.
Some of them apparently are at 500°F [1].
Wouldn't it be a lot more feasible to just make the datacenter on the shore and make a liquid coolant loop that cools in the water? Like the Toronto office towers.
Earlier quoted context omitted.
For a, you can put a sacrificial anode on the frame. Given that the rate of corrosion should be known, the size of it can be calculated to however long the pod is expected to be in service.
This sounds really interesting, can you explain it further for a layman? (Not being snarky, I genuinely know very little about chemistry)
I'm all for exploring new approaches, however: a) Salt water is highly corrosive. Wouldn't maintenance costs be high? b) Isn't marine biology highly sensitive to heat pollution? For high latency services like Amazon Glacier, wouldn't it make sense to host in a place like Iceland? Really cheap hydrothermic/clean power. Highly educated local talent pool, and relatively consistent cool temperatures. If you're maintainin…
1) In the original article, these server farms would be remote administration only.
2) Such facilities would make the secret construction of one's James Bond Villain underwater base much, much easier. (Wouldn't Elon Musk, Jeff Bezos, and Peter Thiel all make great Bond villains?)
This is ridiculous. Data centers contain thousands of racks like the single rack shown in the demo "pod" and they all require hands-on human maintenance from time to time. This becomes quite a bit harder when your server rack is in a sealed steel tube on the seafloor. Splitting the data center into thousands of "pods" will also greatly increase the cost of cabling and connectors for power and data distribution (we're…
Earlier quoted context omitted.
But I require a datacenter with a hull speed of at least 7.5 knots.
let me double check the hull speed formula for datacenters.
Edit: I'm very silly, confusing a few thoughts: e.g natural gas > hydrogen conversion in a cell, assuming some storage of the hydrogen, and less dense non salt water from that - all reducing the pod density but hardly exhausts one traps like a submarine vents.
What would please my errant mind, growing up by the sea, and startled at effluent pipes i could never believe were far enough out, at low tide, would be biomass conversion fuel cells, attached to sewage pipes otherwise dumping into the sea, but i don't know if treatment renders the mass useless as a energy source. The vision in my head was of container pods popping up, floated from their submersion, for maintenance, kinda being hooked in like exhausted tuna, winched aboard a maintenance vessel once a year, for a data catch... or being fed denser storage or servers...
However, what this might do well for, is submerging compute jobs that are expected to run for a long time without much or any external communication. Are there any simulation jobs where enough compute fits inside such a container, so requires mo networking to other containers, and can be retrieved when job complete, to gather results? Would the energy savings make any sense from a cooling pov, if a hypothetical internal power source could last say a year or two?
I'm all for exploring new approaches, however: a) Salt water is highly corrosive. Wouldn't maintenance costs be high? b) Isn't marine biology highly sensitive to heat pollution? For high latency services like Amazon Glacier, wouldn't it make sense to host in a place like Iceland? Really cheap hydrothermic/clean power. Highly educated local talent pool, and relatively consistent cool temperatures. If you're maintainin…
I'm all for exploring new approaches, however: a) Salt water is highly corrosive. Wouldn't maintenance costs be high? b) Isn't marine biology highly sensitive to heat pollution? For high latency services like Amazon Glacier, wouldn't it make sense to host in a place like Iceland? Really cheap hydrothermic/clean power. Highly educated local talent pool, and relatively consistent cool temperatures. If you're maintainin…
> b) Isn't marine biology highly sensitive to heat pollution? Impact on the marine environment is briefly mentioned in the last couple of paragraphs of the article, although they're a little dismissive and don't provide any evidence supporting their claim of negligible impact.