SLA-Constrained Carbon-Aware Routing in Geo-Distributed Serverless Clouds

2026-07-24

AI Summary

This work addresses the high carbon emissions resulting from traditional cloud routing strategies that optimize solely for latency while ignoring the spatiotemporal variability of grid carbon intensity. We propose the first carbon-aware serverless routing approach that achieves optimal carbon reduction within the feasible region defined by service-level agreements (SLAs). By integrating real-world carbon intensity data, we formulate a delay-constrained optimization model and design a low-overhead, geographically distributed routing algorithm that dynamically directs requests to low-carbon regions without compromising user experience. Experiments across five AWS primary regions demonstrate up to 46.8% carbon reduction and an average of 27.4% under mixed workloads, with zero SLA violations and routing overhead accounting for less than 0.02% of total latency. Extending to twelve regions yields an average reduction of 47.5%, effectively supporting Sustainable Development Goals 7 and 13.

Abstract

Modern cloud deployments distribute applications across multiple geographic regions, yet standard routing mechanisms prioritize latency while ignoring the fluctuating carbon intensity of local power grids. Latency-driven routing incurs avoidable carbon emissions, particularly when cleaner regions are within acceptable latency bounds. The proposed model formulates the carbon-aware serverless routing problem as a constrained optimization over geo-distributed cloud regions and introduces an SLA-constrained carbon-aware routing policy that achieves optimal carbon reduction within the SLA-feasible region, evaluated using real carbon intensity measurements across 5 primary AWS deployments. Experimental results show that the proposed policy achieves up to 46.8% carbon reduction while maintaining zero SLA violations across all evaluated thresholds. The system reduces carbon by an average of 27.4% under mixed workloads, and the routing overhead is very low (less than 0.02% of total request latency). A scalability study across 12 AWS regions spanning 6 continents demonstrates that average carbon savings increase from 27.4% to 47.5% as routing flexibility expands under mixed workloads. The proposed work contributes to SDG 13 (Climate Action) and SDG 7 (Affordable and Clean Energy) by enabling low-carbon routing decisions. These results indicate that cloud systems can achieve significant carbon savings without compromising user experience.

Problem

• carbon-aware routing
• geo-distributed clouds
• SLA constraints
• serverless computing
• carbon intensity

Innovation

• carbon-aware routing
• SLA-constrained optimization
• geo-distributed serverless
• sustainable cloud computing
• carbon intensity