Using Information Theory Principles to Schedule Real-time Tasks

The purpose of this paper is to present a scheduling solution based on information theory principles to schedule real-time tasks. We propose the mathematical background for using information as a parameter in real-time systems as well as the relationship between information and utilization. We present a new dynamic priority scheduling solution that selects the task with the highest amount of information per studied interval. We propose the feasibility analysis of the scheduling solution and we compare its performance against the Earliest Deadline First (EDF) scheduling algorithm using as dependent variables the number of context switches and the number of preemptions. We generated 16 test files (with 100 synthetic task sets per tile) using as independent variables: (a) Utilization (from 70% to 100%), (b) Number of tasks per Task set (from 2 to 5) and (e) Hyper-period (fixed at 40). The results showed that: (a) Our scheduling solution improves the performance of EDF by 1.1384% in terms of the number of context switches and 2.0428% in terms of the number of preemptions for the studied task sets; (h) The similarity ratio (number of similar schedules) between the two algorithms was 42.68%.