Optimizing CPU task scheduling is crucial for increasing system performance, particularly in multi-core environments. Traditional algorithms, such as first-come, firstserved (FCFS), shortest-job-first (SJF), and Round Robin (RR), have become inefficient and fail to efficiently reduce waiting times or adapt to changes in the system. To address these issues, we introduce the Adaptive Mountain Gazelle Optimizer (AMGO), a modified version of the Mountain Gazelle Optimizer. It has the ability to dynamically balance exploration and exploitation, assigning tasks according to expected processing times, priorities, and deadlines. Its performance was compared with metaheuristic algorithms such as PSO, GWO, and MGO, as well as traditional algorithms. The results showed that the IMGO reduced waiting time by 30.6%, improved makespan completion time by 9.9%, and reduced total cost by 23.1% when examined on 150 tasks distributed across four cores over 100 iterations. These results demonstrate how IMGO improves real-time task scheduling performance. It has become a contemporary alternative to traditional methods in advanced computing systems for its ability to achieve a balance between exploration and exploitation.
Ahmed et al. (Wed,) studied this question.