ORR: Optimized Round Robin CPU Scheduling Algorithm

The time-specific applications are assigned to Central Processing Unit (CPU) of the system and one of the most promising functions of the time-sharing operating systems is to schedule the process in such a way that it gets executed in minimal time. At present, the Round Robin Scheduling Algorithm (RRSA) is the most widely used technique in a timesharing operating system because it gives better performance than other scheduling techniques, namely, First Come First Serve (FCFS), Shortest Job First (SJF), and Priority scheduling. The major challenge in RRSA is the static value of Time Quantum (TQ) which have plays a pivotal to decrease or increase the performance of the system. In existing literature, many statistical techniques are used for identifying efficient time quantum for RRSA. However, there is limited exposure in existing literature on generating a learning model for identifying optimized TQ. In this research work, a new research direction is given for identifying Optimized TQ by training a learning model and predicting optimum TQ value.

Thus, a new Optimized Round Robin (ORR) CPU Scheduling Algorithm is proposed for time-sharing operating systems by generating the knowledge base of feature set. The ORR is experimentally compared with RRSA and five other improved versions of RRSA. The experimental results show that ORR outperforms in terms of minimizing the Average Waiting Time (AWT), Average Turnaround Time (ATAT)

Number of Context Switch (NCS) and maximizing the throughput of the system.