A Parallelizing Algorithm for Real-Time Tasks of Directed Acyclic Graphs Model

Manar Qamhieh (1) , Serge Midonnet (1) , Laurent George (2)

1 LIGM - Laboratoire d'Informatique Gaspard-Monge
2 AOSTE - Models and methods of analysis and optimization for systems with real-time and embedding constraints

In this paper, we consider parallel real-time tasks follow- ing a Directed Acyclic Graph (DAG) model. This task model is classical in embedded and industrial system applications. Each real-time task is defined by a set of subtasks under precedence constraints. With each subtask being associated a worst case execution time and a maximal degree of parallelism. We propose a parallelizing algorithm based on the critical path concept, in which we find the best parallelizing structure of the task according to the response time and the required number of processors, considering the worst case execution time of the subtasks.

Domaines: Algorithme et structure de données [cs.DS]

Manar Qamhieh, Serge Midonnet, Laurent George. A Parallelizing Algorithm for Real-Time Tasks of Directed Acyclic Graphs Model. RTAS'12 : The 18th IEEE Real-Time and Embedded Technology and Applications Symposium. Work-In-Progress Session, Apr 2012, Beijing, China. pp.45-48. ⟨hal-00695818⟩