Joint Energy Management and Coordinated AIGC Workload Scheduling for Distributed Data Centers: A Diffusion-Aided Reward Shaping Approach

2026-05-03

AI Summary

This work addresses the challenges of high energy consumption and scheduling complexity in AIGC services deployed across distributed data centers, stemming from model heterogeneity, implicit quality-of-service (QoS) evaluation, and intricate inference pipelines. To tackle these issues, the authors propose a joint energy management and collaborative workload scheduling framework that explicitly models QoS to enable cross-provider task migration and fine-grained inference configuration, while integrating diverse energy resources to enhance power flexibility. A key innovation is the introduction of a diffusion model–assisted reward shaping mechanism, which generates complementary multi-step denoising-based reward signals to alleviate reward sparsity in deep reinforcement learning. Experimental results on real-world models and datasets demonstrate that the proposed approach significantly outperforms baseline methods, achieving faster policy convergence and higher system utility while effectively adapting to dynamic electricity pricing and model heterogeneity.

Abstract

Artificial intelligence-generated content (AIGC) has emerged as a transformative paradigm for automating the creation of diverse and customized content, giving rise to rapidly growing computational workloads in cloud data centers. It is imperative for AIGC service providers (ASPs) to strategically schedule AIGC workloads to reduce data center energy costs while guaranteeing high-quality content generation. However, the distinctive characteristics of AIGC services pose critical challenges, including model heterogeneity across ASPs, implicit service quality evaluation, and complex inference process control. To tackle these challenges, we propose a joint energy management and coordinated AIGC workload scheduling framework, which introduces an explicit mathematical characterization of service quality to promote both job transfer among ASPs and fine-grained inference process configuration. Moreover, various energy resources within data centers are jointly considered to enhance power usage flexibility. Subsequently, a system utility maximization problem is formulated to balance AIGC service revenue with operational penalties and costs. Nevertheless, the strong coupling among job scheduling decisions induces severe reward sparsity, which limits the effectiveness of existing deep reinforcement learning (DRL) algorithms. To address this issue, we develop a diffusion model-aided reward shaping approach to synthesize complementary reward signals through a multi-step denoising process. This approach is seamlessly integrated with DRL to enable efficient learning of scheduling policies under sparse environmental feedback. Experiments based on real-world models and datasets demonstrate that our scheme effectively accommodates electricity price fluctuations and AIGC model heterogeneity, while achieving superior learning convergence and system utility compared with benchmark methods.

Problem

• AIGC
• workload scheduling
• energy management
• service quality
• distributed data centers

Innovation

• AIGC workload scheduling
• reward shaping
• diffusion model
• energy management
• deep reinforcement learning