CarbonCP: Carbon-Aware DNN Partitioning with Conformal Prediction for Sustainable Edge Intelligence
Preprint
Abstract
This paper presents a solution to address carbon emission mitigation for end-to-end edge computing systems, including the computing at battery-powered edge devices and servers, as well as the communications between them. We design and implement, CarbonCP, a context-adaptive, carbon-aware, and uncertainty-aware AI inference framework built upon conformal prediction theory, which balances operational carbon emissions, end-to-end latency, and battery consumption of edge devices through DNN partitioning under varying system processing contexts and carbon intensity. Our experimental results demonstrate that CarbonCP is effective in substantially reducing operational carbon emissions, up to 58.8%, while maintaining key user-centric performance metrics with only 9.9% error rate.Keywords
All Related Versions
This publication has 4 references indexed in Scilit:
- An adaptive DNN inference acceleration framework with end–edge–cloud collaborative computingFuture Generation Computer Systems, 2022
- Algorithmic Learning in a Random World2022
- The real climate and transformative impact of ICT: A critique of estimates, trends, and regulationsPatterns, 2021
- Edge AI: On-Demand Accelerating Deep Neural Network Inference via Edge ComputingIEEE Transactions on Wireless Communications, 2020