A Systematic Review on Task Scheduling in Cloud Computing

Abstract

Cloud computing provides on-demand access to virtualized resources, enabling efficient task execution and service delivery. Among its core components, task scheduling plays a crucial role in optimizing resource utilization, reducing execution time, and enhancing overall system performance. This systematic literature review examines 58 research articles published between 2018 and June 2025, focusing on scheduling techniques in cloud computing. The review categorizes existing approaches—including traditional, machine learning-based, heuristic, metaheuristic, and hybrid methods—while analyzing their objectives, considered factors, and reported limitations. Findings reveal that although significant advancements have been achieved, key challenges remain in ensuring availability, reliability, adaptability, and energy efficiency, particularly under dynamic and heterogeneous cloud environments. Many proposed methods demonstrate strong results in controlled experiments but struggle with scalability in real-world deployments. Moreover, issues such as uncertainty handling, SLA compliance, fault tolerance, and sustainability are insufficiently addressed. The review highlights the adaptability of existing algorithms and their potential for integration with emerging technologies like artificial intelligence, edge computing, and the Internet of Things (IoT). Future research opportunities lie in developing intelligent, multi-objective, and context-aware scheduling strategies that balance cost, performance, energy consumption, and security. Overall, task scheduling in cloud computing remains a complex yet promising area of study, with significant potential for innovation to meet evolving user and system demands.