Combining EDF and LST to enhance the performance of real-time task scheduling

Real time system is a system in which the computation not only depends on producing a correct output, but the output should be delivered within predetermine timing constraint called deadline. As humans, we handle scheduling issues every day. For example, a student must accomplish homework by the appropriate due date, a professor must complete the rough draft of a paper by the submission date, etc., where we have only one task to accomplish, meeting that deadline probably would be very simple. Our lives, however, contain many tasks that have deadlines tax forms, car inspections, meetings, classes, etc. Thus, we must use some sort of scheduling technique to “juggle” our various tasks, so that they all are completed by their appropriate deadlines. As a result, many scholars develop different static and dynamic scheduling algorithms, which are best suited for both uniprocessor and multiprocessors. This paper work will focus on proposing a new dynamic scheduling algorithm. Because, it is found that the well-known uniprocessor dynamic scheduling (EDF (earlier deadline first) and LLF (least slack time first)) algorithms have many context switching and many probability of overhead occurrence (deadline-tie and laxity-tie respectively) which will degrade the performance of the whole system. The paper proposes a dynamic scheduling algorithm that has low context switching and totally reduce the probability of overhead occurrence by combining the deadline and slack to give priority for each task in the ready queue.