A starvation free IMLFQ scheduling algorithm based on neural network

Scheduling algorithms are used in operating systems to optimize the usage of processors. One of the most efficient scheduling algorithms called Multi-Layer Feedback Queue (MLFQ) uses several queues with different quanta. The most important weakness of this method is the inability to define the optimized number of the queues and quantum of each queue also this algorithm has starvation for some process. These factors affect the response time directly. Also this algorithm does not show any considerable improvement in response time of the processes in comparison with the other scheduling algorithms. In this paper, a new algorithm is presented to solve these problems and minimize the response time simultaneously. Also we proposed a new idea to solve the starvation problem while considering the processes priority too. In this algorithm Recurrent Neural Network has been utilized to find both the number of queues and the optimized quantum of each queue. Also in order to prevent any probable faults in processes' response time computation, a new fault tolerant approach has been presented. The experimental results show that using the IMLFQ algorithm results in better response and waiting time in comparison with other scheduling algorithms. Experimental results show that the proposed method decreases the response time of the time consuming processes and also resolves the starvation problem.