Memory and time-efficient schedulability analysis of task sets with stochastic execution time

This paper presents an efficient way to analyse the performance of task sets, where the task execution time is specified as a generalized continuous probability distribution. We consider fixed task sets of periodic, possibly dependent, non-pre-emptable tasks with deadlines less than or equal to the period. Our method is not restricted to any specific scheduling policy and supports policies with both dynamic and static priorities. An algorithm to construct the underlying stochastic process in a memory and time efficient way is presented. We discuss the impact of various parameters on complexity, in terms of analysis time and required memory. Experimental results show the efficiency of the proposed approach.