QoS provisioning of a task-scheduling algorithm for lightweight devices

Abstract

Task scheduling is a main objective in operating system design. Many operating systems have been designed and applied to various systems. With the upcoming Internet of things, the requirements for the operating system will vary according to the application. We propose a round-robin-based scheduling algorithm to satisfy the quality of service (QoS) of real-time tasks. Many conventional scheduling algorithms have dealt with the real-time issue, but they still require improvement. Moreover, most assume the burst time is known beforehand, an unrealistic assumption in practical implementations. Thus, we present a heuristic method to predict the burst time of each task and a round-robin-based weighted scheduling algorithm providing QoS to real-time tasks for lightweight devices. The performance of the proposed scheduling algorithm is verified on an embedded system. Experimental results proved the QoS of this algorithm was satisfied.