POA: AI-Driven Meta-Heuristic Scheduling Algorithm for Enhancing Resource Utilization in Cloud Environment
C N Veena, Ravindra Eklarker, Kaushik Shivakumar, V. Latha, Rani Sailaja Velamakanni, Santhosh Kumar Medishetti
Abstract
Cloud Computing (CC) has become an essential platform for executing complex and large-scale tasks by providing on-demand resources and services. Efficient Task Scheduling (TS) plays a vital role in ensuring optimal resource utilization, reduced energy consumption, minimized execution time, and overall cost reduction. However, traditional scheduling algorithms often struggle with balancing exploration and exploitation, leading to issues like task delays and resource underutilization. To address these challenges, this paper proposes a novel task scheduling approach using the Pufferfish Optimization Algorithm (POA). Inspired by the self-inflation and defence mechanism of pufferfish, POA effectively balances exploration and exploitation by dynamically adjusting its search patterns in response to environmental changes. The proposed algorithm focuses on optimizing key parameters such as makespan, energy consumption, and resource utilization while minimizing operational costs. Simulation results demonstrate that POA outperforms existing metaheuristic algorithms such as PSO, GA, and ACO, achieving a 23.5% reduction in energy consumption, 19.8% improvement in resource utilization, 17.4% reduction in operational cost, and 21.7% reduction in makespan. This work highlights the potential of POA as an intelligent and adaptive solution for efficient task scheduling in cloud environments, especially for time-sensitive and large-scale applications.