Cloud Continuum systems are inherently complex and massive, often featuring disparate infrastructure stakeholders, heterogeneous platforms, and distributed energy providers; this significantly increases the complexity of developing, deploying, and managing applications. The Serverless computing offers a powerful tool to simplify and speed up the Continuum application development. However, existing scheduling mechanisms for Serverless platforms primarily focus on performance metrics such as latency, model accuracy, and throughput, often neglecting critical factors such as energy efficiency and sustainability. This gap is further exacerbated in Continuum environments, where computational nodes may rely on unpredictable and intermittent green energy sources, leading to availability bottlenecks and energy constraints.

This work explores the opportunity to design a decentralized energy management approach for scheduling Serverless functions in the Cloud Continuum. To achieve this, we formally model a green energy-aware workload scheduling problem incorporating the variability of renewable energy nodes and the QoS requirements of Serverless functions. Then, we develop a decentralized scheduling mechanism based on multi-agent reinforcement learning that leverages distributed agents to consider the energy awareness into scheduling decisions, prioritizing nodes powered by renewable energy while ensuring task completion within required QoS constraints. To demonstrate the practicality of our approach, we develop a real-world prototype using cluster of Raspberry Pis enabled with multi PiAgent modules, Cloud servers, Lightweight Kubernetes, and OpenFaaS. The experimental results show that the proposed approach maximizes the green energy utilization by (44%) while reducing the total latency by (25%) compared to the centralized technique, demonstrating its energy efficiency, scalability, and overall sustainability in Continuum settings.