An Attempt to Set Standards for Studying and Comparing the Efficiency of Round Robin Algorithms

With the advent of the need for interactive systems, the urgent need for time-sharing systems has emerged. Round-robin algorithms have emerged to achieve time-sharing. The degree of performance of time-sharing systems depends largely on the length of the time slice in the round-robin algorithms. The length of the time slice affects the measuring criteria of the performance of the algorithms. Researchers suggested and are continuing suggesting algorithms in order to obtain the best values for the time slice. Adopting one algorithm over another in a system and for a class of applications requires choosing the best performing algorithm. This research is an attempt to develop an objective approach for accurate comparison between algorithms. For the sake of objectivity in comparison, five algorithms similar in their general characteristics were chosen; Modified Median Round Robin Algorithm(MMRRA), A New Median-Average Round Robin Scheduling Algorithm(NMARR), An Improved Round Robin Scheduling Algorithm with Varying Time Quantum (IRRVQ), A Modified Round Robin CPU Scheduling Algorithm with Dynamic Time Quantum (RRDT), Improved Round Robin Algorithm with Progressive Dynamic Quantum (IRRPDQ). The results showed that the outperformance of an algorithm over a group of algorithms according to a specific criterion is not permanent and fixed in value, and that resorting to statistical measures is the best way to clarify the degree of performance of the algorithms.