Move the AI workload to the energy: energy-aware scheduling on Akash
Solving the AI energy crisis | Greg Osuri on what it takes to power AI (Changelog)
“Why burn energy but instead use that energy to train AI? Where it’s windy in Kansas, move the AI workload there. When it’s sunny in California, move the AI workload. The AI workload [has] to be elastic and has to be asynchronous… Akash, we are developing an energy-aware [scheduler] that will pick the lowest energy. So we’ll start advertising the cost per energy now.” — 00:46:49
Context: Answering whether plugging GPUs into wasted renewable capacity solves the problem (“Absolutely”); he cites wind at 1 cent/kWh vs LNG at 10 cents and diesel at 32 cents.