About The process involves measuring the carbon emissions generated from deep learning training. If a server's training process produces high carbon emissions, we transfer the workload to servers with lower carbon emissions. Ultimately, this method aims to achieve a carbon net-zero state.
deep-learning
gpu-computing
carbon
dvfs
carbon-footprint
model-predictive-control
monitoring-systems
stramlit
net-zero
carbon-friendly
moving-between-clouds
-
Updated
Mar 5, 2024 - Python