Downloads: 0
Research Paper | Computer Science | Volume 15 Issue 5, May 2026 | Pages: 1222 - 1228 | United States
Energy-Efficient Kubernetes Scheduling for Green Cloud Computing: A Carbon-Aware Workload Placement Framework
Abstract: Data center energy consumption attributable to containerized cloud workloads has grown substantially with the adoption of Kubernetes at enterprise scale, yet Kubernetes scheduling decisions remain entirely agnostic to the energy and carbon implications of workload placement. This paper presents CASK (Carbon-Aware Scheduler for Kubernetes), a novel scheduling framework that extends Kubernetes' standard scheduler with carbon intensity awareness, renewable energy availability signals, and hardware energy efficiency profiles to minimize the carbon footprint of containerized workloads without degrading performance SLAs. CASK integrates real-time carbon intensity data from regional electricity grid APIs, hardware thermal design power (TDP) profiles for heterogeneous node fleets, and time-of-day renewable availability forecasts into a multi-objective scheduling function that balances carbon minimization with latency and availability constraints. Evaluation across 3,200 nodes in two hyperscale enterprise Kubernetes deployments over 20 weeks demonstrates that CASK reduces mean carbon emissions per workload unit by 28.3%, shifts 34% of deferrable batch workloads to low-carbon time windows, and achieves energy efficiency improvements of 19.4% through hardware-aware pod placement on energy-efficient node types- without exceeding a 4.2% mean increase in workload completion time for deferrable workloads. A carbon-aware scheduling policy framework, renewable energy signal integration architecture, and Kubernetes scheduler plugin implementation guide are presented.
Keywords: Kubernetes, Green Cloud Computing, Energy Efficiency, Carbon-Aware Scheduling, Sustainable Computing, Container Orchestration, Carbon Footprint
How to Cite?: Dinesh Kumar Movva, "Energy-Efficient Kubernetes Scheduling for Green Cloud Computing: A Carbon-Aware Workload Placement Framework", Volume 15 Issue 5, May 2026, International Journal of Science and Research (IJSR), Pages: 1222-1228, https://www.ijsr.net/getabstract.php?paperid=SR26519073728, DOI: https://dx.dx.doi.org/10.21275/SR26519073728