A heuristic to minimize the cardinality of a real-time task set by automated task clustering

Abstract

International audienceWe propose in this paper a method to automatically map functionalities (blocks of code corresponding to high-level features) with real-time constraints to tasks (or threads). We aim at reducing the number of tasks functions are mapped to, while preserving the schedulability of the initial system. We consider independent tasks running on a single processor. Our approach has been applied with fixed-task or fixed-job priorities assigned in a Deadline Monotonic (DM) or a Earliest Deadline First (EDF) manner

Minimizing a real-time task set through Task Clustering

“…We cannot compare our heuristic with an optimal solution because the task clustering is not achievable with an exhaustive search among all partitions. Instead we compare our results with our previous work [7]. We also present the part of zero-cost clustering performed in the clustering.…”

“…In [7], we gave a practical idea of the high complexity of the task clustering. Our experiments showed that an exhaustive search is not achievable due to the exponential number of partitions to explore.…”

“…We also propose conditions under which tasks can be grouped without affecting the schedulability. This research follows the work we made in [7] where we laid the basis for the task clustering and proposed a first heuristic. In this paper, we mainly develop four improvements.…”

A holistic optimization approach for the synthesis of AUTOSAR E/E architecture

“…In this context, several works have been carried out [1][2][3][4]. Most of them deal with the problem of mapping runnable to task and/or mapping SWCs to ECUs.…”

A resource efficient framework to run automotive embedded software on multi-core ECUs

“…On the other hand, when most of the runnables have different periods, the number of tasks can go as high as above the maximum number of tasks allowed by AUTOSAR OS. To deal with this problem, task merging approaches have emerged to diminish the number of tasks, 180 such as [31,32]. Nonetheless, the merging approaches presented in these papers are not directly applicable to our problem, because they attempt to minimize the number of tasks even if it results in a higher processor utilization, as long as the task set on each core stays schedulable with respect to the scheduling policy in use on the processor.…”