Sarcouncil Journal of Engineering and Computer Sciences

Abstract: The rapid increase in the use of cloud-native applications has increased the environmental footprint of distributed computing, especially in multi-region Kubernetes deployments running on platforms such as Amazon Web Services (AWS). This paper gives a detailed description of how to design and run energy-efficient Kubernetes workloads across geographically separated AWS regions. It considers energy-conscious scheduling, efficiency-optimized autoscaling, telemetry-enabled observability, sustainable software design, and policy-based governance through applying concepts of sustainable computing in concert with container-based orchestration techniques. The article focuses on matching real-time use of resources with real-time environmental conditions (including carbon intensity, as well as regional energy mix) without sacrificing service reliability or compliance. In an integration of new studies and best practices, the research sets the stage for how to approach the process of scalable, resilient, and environmentally responsible Kubernetes operations.