Advanced Multi Level Feedback Queue Scheduling Algorithm using Machine Learning

Process scheduling is crucial in optimizing resource utilization and reducing waiting times in computer systems. The traditional multilevel feedback queue scheduling algorithm (MLFQ) divides processes into multiple queues based on priority and uses round-robin and shortest-job-first algorithms to process them. We have applied the k-means clustering algorithm to divide the processes into three queues based on their remaining time. Processes can move between queues which deals with the problem of both aging and starvation. Moreover, we have used a dynamic quantum time round-robin scheduling algorithm in all three queues, where the quantum time for a queue is the multiplication of the median of the remaining time of processes in the queue by the priority of that queue. This paper proposes a novel scheduling algorithm that combines k-means clustering with multilevel feedback queue (MLFQ) scheduling to optimize resource utilization, minimize waiting time, and maximize throughput in process scheduling systems. By addressing the challenges of aging and starvation through dynamic quantum times and utilizing multiple queues, this algorithm aims to mitigate these issues while providing a sustainable solution.