Joint Energy Management and Coordinated AIGC Workload Scheduling for Distributed Data Centers: A Diffusion-Aided Reward Shaping Approach
arXiv:2605.02965v2 Announce Type: replace 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 se
The rapid growth of AIGC workloads is pushing existing data center infrastructure to its limits, necessitating more efficient energy management and scheduling solutions.
This research addresses the critical challenge of optimizing energy consumption in data centers while meeting the escalating demands of AIGC, directly impacting operational costs and sustainability.
New computational and scheduling paradigms are emerging to manage the unique demands of AIGC, moving towards more intelligent and energy-efficient data center operations.
- · AIGC service providers
- · Cloud data center operators
- · AI infrastructure companies
- · Energy management software developers
- · Inefficient data center operators
- · Legacy energy management systems
Reduced operational expenditures for data centers supporting AIGC.
Increased availability and scalability of AIGC services due to optimized resource allocation.
Potential for breakthroughs in sustainable AI infrastructure, influencing global compute buildout strategies.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG