Authors: Swathi S, Abinaya C S
Abstract: The rapid growth of cloud computing has led to a significant increase in energy consumption and carbon emissions from large-scale data centers, making sustainability a critical concern. Traditional cloud resource management techniques primarily focus on optimizing performance and cost, often neglecting the environmental impact of cloud operations. Existing green cloud solutions mainly emphasize energy efficiency; however, energy-aware approaches alone are insufficient, as carbon emissions vary dynamically with the carbon intensity of the power grid. This paper addresses these limitations by proposing a hybrid carbon- and energy-aware scheduling approach for green cloud computing. The proposed solution dynamically allocates and migrates cloud workloads based on real-time energy consumption, carbon intensity, and Service Level Objective (SLO) constraints. By integrating workload classification and SLO-aware decision-making, the system ensures reduced environmental impact without compromising application performance. The proposed approach aims to achieve substantial reductions in carbon emissions and overall energy usage while maintaining acceptable latency and quality of service. This work contributes toward building sustainable, environmentally responsible cloud infrastructures suitable for next-generation cloud data centers.
