Schedulability analysis for independent partitions in integrated modular avionics systems

Abstract

Recently the integrated modular avionics (IMA) architecture has been widely adopted by the avionics industry due to its strong partition mechanism. Although the IMA architecture can achieve effective cost reduction and reliability enhancement in the development of avionics systems, it results in a complex allocation and scheduling problem. In this paper, we analyze the schedulability of independent partitions and provide proper schedules for multi-core IMA systems. We first calculate the maximum scaling factor, by which the execution times of all partitions can be multiplied without violating the timing requirements of the systems. Then, we use the scaling factor obtained to determine whether all partitions are schedulable and produce a valid allocation for each partition. Finally, simulation experiments are conducted to show the efficiency and reliability of our approach in terms of time consumption and acceptance ratio.

A load balancing method for avionics systems via artificial bee colony and simulated annealing algorithms

“…The inefficient allocation technology is faced with the challenge of over-utilization and underutilization (unbalanced) of resources, which results in the decline of service performance (in the case of over-utilization) or the waste of resources (in the case of under-utilization) [2,3,12]. The basic idea of resource allocation in avionics systems is to allocate tasks (diversity and complexity) between resources by way of tailored algorithms to avoid load balancing problems [13], so as to achieve the best control effect [14]. Load balancing algorithms should also optimize key performance parameters, such as response time, completion time, reliability, availability, energy consumption, cost, resource utilization, etc [15,16].…”