Static Scheduling in Real Time for Resource Optimization

Abstract

Scheduling of tasks on the system is a decision making process which deals with performing functions like allocation of common resources, instruction sequencing, etc at specified intervals. Real time systems have become inseparable part of our lives and find many applications indifferent areas. The correctness of the results in real time systems depends not only on the logical computations but also on the time at which results are produced. A system needs to perform different tasks and scheduling of these tasks is done on the processor. It helps in the execution and completion of every task within desired time. For the efficient working of real time system it is necessary that all the tasks should meet their deadlines. The real time tasks can be broadly classified into three categories on the basis of deadline. The tasks which result in catastrophic conditions on missing the deadline are the hard real time tasks. The tasks whose results produced after missing the deadlines are of no use are known as firm real time tasks and those tasks whose results are useful even after missing the deadline are known as soft real time systems. The pre-emptions allowed in the real time systems and the overhead increases as the number of preemptions increases. Most of the embedded real-time systems only use necessary resources so that the extra pre-emption overheads among tasks should not degrade the system performance. The priority assignment schemes assign priorities to different tasks which can be static or dynamic. The study of various scheduling algorithms along with the comparison between different real time scheduling algorithms has been done in the thesis. In order to find a feasible schedule, a schedulability test is required which should ensure the meeting of deadlines. Resource optimization is of main concern in case of scheduling in real time. Shortest path algorithms achieve resource optimization by finding the shortest path between different nodes. Shortest distance between the nodes should be optimally evaluated which results in resource optimization and the tasks are scheduled through that optimized path in order to achieve feasible and efficient schedule.