Improved Initialization for Tunicate Swarm Optimization in Cloud Task Scheduling

Efficient task scheduling in cloud computing is critical for optimizing resource utilization and reducing execution cost. However, the task scheduling problem is known to be NP-hard, motivating the use of metaheuristic algorithms. This paper proposes an improved initialization approach for the Tuni-cate Swarm Optimization (TSO) algorithm, leveraging strategic heuristics. Specifically, Shortest Job First Priority (SJFP) and Earliest Completion Time (ECT) heuristics are employed to initialize TSO. Both of the proposed SJFP-TSO and ECT-TSO algorithms are evaluated on the performance metrices including minimizing makespan, total execution time, load balancing, and energy consumption. Comparative analysis shows the superiority of the proposed algorithms over conventional TSO method.