Carbon Intensity-Aware Adaptive Inference of DNNs
This addresses the environmental impact of AI for service providers, but it is incremental as it builds on existing adaptive inference methods.
The paper tackles the problem of reducing the carbon footprint of DNN inference by adapting model size and accuracy based on varying carbon intensity throughout the day, resulting in up to 80% improvement in carbon emission efficiency for vision recognition services.
DNN inference, known for its significant energy consumption and the resulting high carbon footprint, can be made more sustainable by adapting model size and accuracy to the varying carbon intensity throughout the day. Our heuristic algorithm uses larger, high-accuracy models during low-intensity periods and smaller, lower-accuracy ones during high-intensity periods. We also introduce a metric, carbon-emission efficiency, which quantitatively measures the efficacy of adaptive model selection in terms of carbon footprint. The evaluation showed that the proposed approach could improve the carbon emission efficiency in improving the accuracy of vision recognition services by up to 80%.