Optimizing Inter-Core Data-Propagation Delays in Industrial Embedded Systems under Partitioned Scheduling

This paper addresses the scheduling of industrial time-critical applications on multi-core embedded systems. A novel scheduling technique under partitioned scheduling is proposed that minimizes inter-core data-propagation delays between tasks that are activated with different periods. The proposed technique is based on the read-execute-write model for the execution of tasks to guarantee temporal isolation when accessing the shared resources. A Constraint Programming formulation is presented to find the schedule for each core. Evaluations are preformed to assess the scalability as well as the resulting schedulability ratio, which is still 18% for two cores that are both utilized 90%. Furthermore, an automotive industrial case study is performed to demonstrate the applicability of the proposed technique to industrial systems. The case study also presents a comparative evaluation of the schedules generated by (i) the proposed technique and (ii) the Rubus-ICE industrial tool suite with respect to jitter, inter-core data-propagation delays and their impact on data age of task chains that span multiple cores.

1 INTRODUCTION

The requirement for high computational power to support compute-intensive features in embedded system applications has drastically increased in the past few years [17]. To meet this requirement, the developers of these applications have started to employ multi-core architectures as opposed to the contemporary single-core architectures. However, the transition from single-core to multi-core is not trivial, especially in the case of real-time embedded systems, which have stringent timing requirements, e.g., timing constraints on the response times of tasks and data-propagation delays through chains of tasks [3], [5], [16], [21]. In the case of multi-core real-time embedded systems, the interference introduced due to the contention for shared resources such as the system bus and memories can lead to potentially unbounded delays [13], [31], [36]. Schedulability analysis techniques are used to ascertain if the specified timing constraints are met or not. Real-time scheduling and schedulability analysis for single-core real-time embedded systems are well developed and well adopted by the industry [10], [28]. Whereas, these techniques for multi-core real-time systems are still evolving [18].