A comparative evaluation of heuristic and metaheuristic job scheduling algorithms for optimized resource management in cloud environments

An essential element of cloud computing is effective job scheduling, which optimizes resource utilization and minimizes operational costs. Numerous scheduling methodologies have been developed, each targeting specific performance objectives such as reducing makespan, minimizing flow time, decreasing deadline violations, and optimizing resource utilization. The selection of an appropriate scheduling algorithm is crucial for ensuring optimal performance, scalability, and resource efficiency as cloud environments become increasingly complex and dynamic. This study provides a comprehensive analysis and comparison of six prevalent scheduling algorithms, namely FCFS, SJF, LJF, EDF, Max-Min, and PSO, under varying cloudlet loads of 200,400,600,800, and 1000. The CloudSim simulator is applied to evaluate each algorithm using key performance metrics, including makespan, average flow time, and the number of cloudlets that fail to meet deadlines. The results indicate that while SJF excels in minimizing average flow time under lighter workloads, PSO consistently outperforms the other algorithms under heavier loads, demonstrating superior scalability and efficiency in large-scale environments. Although EDF proves effective for time-sensitive tasks, Max Min serves as a robust alternative for fair resource distribution. Overall, this study provides valuable insights for cloud service providers to enhance resource management and improve system performance by emphasizing the importance of algorithm selection based on workload characteristics and scheduling constraints.