An Improved Scheduling Algorithms Of RMS And EDF And Mixed Scheduling Algorithm Of The Two

In the environment of uniprocessor, we discuss the algorithms of RMS which stands for rate-monotonic-scheduling and DEF which stands for the earliest-deadline-first-scheduling detailed, and a mixed scheduling algorithm which is combinations of the rate-monotonic scheduling algorithm and the deadline driven scheduling algorithm. We conclude that, along with the algorithm of RMS which is an optimal fixed priority order assignment algorithm, we can obtain utilization of processor about 70%( for a large task set ); whereas along with a dynamic priority order assignment algorithm of EDF, we can obtain the utilization of processor 100%. At the same time, we also discuss the mixed algorithm which is combinations of the rate-monotonic scheduling algorithm and the deadline driven scheduling algorithm. Here, I want to say that the mixed algorithm is practical as well, and more importantly, we can get a high level utilization of processor. How about the algorithms of RMS and EDF, if they are used in the multiprocessor's environment? In general, we can not say that the algorithm is good or the other is bad and the different algorithm may be used in the different environment in which it can achieve some good results. In the end, we focus on the problem, which is mentioned formerly.The algorithm of RMS is a fixed priority scheduling algorithm. It assigns priorities to tasks according to their request rates. The more frequency the task is, the higher priority the task is. In the long run, it fits for some small systems. On the one hand, the algorithm of RMS is simple and has a little amount of computation, and on the other hand, it may be simple for realization in some environment. The scheduling algorithm which assigns priorities to tasks in a monotonic relation to their request rates was shown to be optimum among the class of all fixed priority scheduling algorithm.A task will be assigned the highest priority if the deadline of its current request is the nearest, and will be assigned the lowest priority if the deadline of its current request is the furthest. In this section, we establish a necessary and sufficient condition for the algorithm of EDF. The dynamic deadline driven scheduling algorithm was then shown to be globally optimum and capable of achieving full processor utilization.The mixed scheduling algorithm combines the algorithms of RMS and EDF, and makes full use of them. In some places, it is proved to be efficiency and effectiveness as well. But in fact, we investigate few aspects about the mixed scheduling algorithm. In this section, we also establish a necessary and sufficient condition for the algorithm of the mixed scheduling.All of the above algorithms are fit for the environment of uniprocessor. Nowadays, with the development of technology of science, the uniprocessor cannot be capable for the need of our society, which motivates us to figure out more algorithms. In this article, I also propose some algorithms and problems which we encountered in the multiprocessor environment and discuss it little. As time went on we want to make it as our stress. Two major strategies can be used to schedule real-time tasks on a multiprocessor: global scheduling or partitioning. In a global scheduling strategy, all task instances are stored in a single queue and, at any given time, the processors run the instances with the highest priority. A partitioning strategy involves two algorithms: One assigns tasks to processors and the other one schedules tasks on each processor. Partitioning strategies have the advantage of reducing the multiprocessor scheduling problem to scheduling problems on individual processors for which many results are known.