A dynamic scheduling algorithm with minimum context switches for spacecraft avionics systems

Abstract

Most of the process-scheduling algorithms, which can be exploited by spacecraft avionics systems, fall into one of the two main categories: static algorithms and dynamic algorithms. Static algorithms assign priorities to processes before runtime. These algorithms bound the CPU utilization and require considerable information about the runtime parameters of processes in advance. Instead, these disadvantages encourage the spacecraft avionics system designers to exploit dynamic algorithms. However, these algorithms have their own disadvantages. For example, these algorithms require so many context switches to schedule the processes. This causes a notable overhead in the runtime. One of the well-known dynamic algorithms is the MLF (Minimum Laxity First) algorithm. The MLF algorithm suffers from a serious problem (in addition to the need for so many context switches). This paper proposes a novel dynamic algorithm called Optimized MLF which is an attempt to solve the problems of the MLF algorithm in order to make it more applicable to spacecraft avionics systems. The performance of the proposed scheduling algorithm is evaluated through the use of mathematical modeling as well as simulation results. Both the mathematical model and simulation results show that the optimized MLF algorithm requires less context switches than the traditional MLF and makes the MLF algorithm more applicable to spacecraft avionics systems.