An Optimized Beluga Whale Approach for Migration to Reduce Power and Service Level Agreement in Real-Time System

Abstract

Managing power consumption in cloud data centers has become a critical challenge. Live container migration is a technology supporting energy efficiency in this context. To address the hurdles of power consumption management, thereby mitigating carbon emissions and minimizing service level agreement (SLA) violations, we propose an approach utilizing a real-time server with the Beluga Whale Optimization Algorithm (BWOA) for container migration. The proposed approach aims to optimize energy consumption while ensuring SLA compliance. The BWOA, a machine learning-based method, is employed to predict the resource requirements of containers and migrate them to hosts with sufficient resources. We implemented the proposed approach in a real-time cloud server and compared its performance with other algorithms in terms of response time. The results demonstrate a remarkable 30% improvement in response time, leading to reduced SLA violations and optimized power consumption in containerized data centers.