SOREL , Yves

In this paper, we address for safety critical ap- plications the problem of multiprocessor real-time scheduling while taking into account the exact preemption cost. In the framework of multiprocessor real-time partitioned scheduling, we propose a greedy heuristic which balances the load of the tasks on all the processors and minimizes the...

We address distributed real-time applications represented by systems of non-preemptive dependent periodic tasks. This system is described by an acyclic directed graph. Because the distribution and the scheduling of these tasks onto a multiprocessor is an NP-hard problem we propose a greedy heuristic to solve it. Our heuristic sequences three...

First we justify our concern in latency constraints for real-time systems with precedence constraints. We evoke the model based on graph theory used to state and solve the problem of non-preemptive scheduling systems with precedence and latency constraints. Because latency constraints involve pairs of operations related to the graph which...

Partial answers have been provided in the real-time literature to the question whether preemptive systems are better than non-preemptive systems. This question has been investigated by many authors according to several points of view and it still remains open. Compared to preemptive real-time scheduling, non-preemptive real-time scheduling and...

Parce qu'un systèmes temps réel combine fonctionnalités de contrôle et traitement de données, il est souvent spécifié à l'aide de plusieurs langages adaptés à ces deux aspects. La plupart de ces systèmes étant aujourd'hui distribués le problème est ensuite d'obtenir une implantation de ces spécifications distinctes. En effet une distribution de...

Most safety critical embedded systems, i.e. systems for which constraints must necessarily be satisfied in order to avoid catastrophic consequences, consist of a set of data dependent tasks which exchange data. Although non-preemptive realtime scheduling is safer than preemptive real-time scheduling in a safety critical context, preemptive...

This paper deals with load balancing and efficient memory usage for homogeneous distributed real-time embedded applications with dependence and strict periodicity constraints. Most of load balancing heuristics tend to minimize the total execution time of distributed applications by equalizing the workloads of processors. In addition, our...

Non-preemptive real-time scheduling and the corresponding schedulability analyses have received considerable less attention in the research community, compared to preemptive real-time scheduling. However, non-preemptive scheduling is widely used in industry, especially in the case of hard real-time systems where missing deadlines leads to...

Most safety critical embedded systems, i.e. systems for which constraints must necessarily be satisfied in order to avoid catastrophic consequences, consist of a set of data dependent tasks which exchange data. Although non-preemptive realtime scheduling is safer than preemptive real-time scheduling in a safety critical context, preemptive...

We start this chapter by explaining how modeling needs of thereal-time community have lead to the introduction of synchronouslanguages in the early 1980s. We present the characteristics of theselanguages as well as their limitations. We then propose aclassification of synchronous languages and of the languages relatedto them. This introduces a...

In this paper we study hard real-time systems composed of independent periodic preemptive tasks where we assume that tasks are scheduled by using Liu & Layland's pioneering model following the Rate Monotonic Analysis (RMA). For such systems, the designer must guarantee that all the deadlines of all the tasks are met, otherwise dramatic...

National audience

We consider the problem of minimizing bus usage for static real-time scheduling of hierarchical dataflow specifications involving conditional execution. Statically scheduling conditional communications over an asynchronous broadcast bus involves the sending of the activation conditions themselves, which allow all processors to know which...

In this paper we study hard real-time systems composed of independent periodic preemptive tasks in the monoprocessor case. For such systems it is mandatory to satisfy all the constraints for all tasks. Although preemptive scheduling algorithms are able to successfully schedule some systems that cannot be scheduled by any non preemptive...

In this paper we provide a schedulability test for dependant periodic task systems scheduled using non-preemptive policies. We consider the case of one processor and the schedulability test is based on the periodicity of a feasible schedule. A feasible schedule is periodic if it repeats from a time instant s with a period p. In the case of one...

Les langages synchrones sont issus de la communauté temps réel, comme l'attestentles premières publications sur le sujet, datant des années 1980. Ils répondent au besoin de spécifier de manière non ambigüe les fonctionnalités de systèmes de commande complexes soumis à des contraintes de temps stictes (dures), appelés aussi systèmes temps réel...

In this report we study hard real-time systems composed of dependent strictly periodic preemptive tasks in the monoprocessor case. Although preemptive scheduling algorithms are able to successfully schedule some systems that cannot be scheduled by any non preemptive scheduling algorithm, the cost of preemption may not be negligible. Therefore,...

Classical approaches based on preemption, such as RM (Rate Monotonic), DM (Deadline Monotonic), EDF (Earliest Deadline First), LLF (Least Laxity First), etc, give schedulability conditions in the case of a single processor, but assume the cost of the preemption to be negligible compared to the duration of each task. Clearly the global cost is...

In this paper, we focus on the semi-partitioned scheduling of sporadic tasks with constrained deadlines and identical processors. We study two cases of semi-partitioning: (i) the case where the worst case execution time (WCET) of a job can be portioned, each portion being executed on a dedicated processor, according to a static pattern of...

In this paper, we focus on the semi-partitioned scheduling of sporadic tasks with constrained deadlines and identical processors. We study two cases of semi-partitioning: (i) the case where the worst case execution time (WCET) of a job can be portioned, each portion being executed on a dedicated processor, according to a static pattern of...

This paper presents a connection between two real-time models: a deadline-based model and a latency-based model. The importance of the latency-based model is proved through a result showing that two deadlines, instead of a latency constraint, over-constrain the real-time applications. Moreover, we give a deadline-marking algorithm based on the...

We consider the problem of fixed priority scheduling of non-preemptive strict periodic tasks in conjunction with sporadic preemptive tasks. There are few studies about the scheduling problem combining these two kinds of tasks. Moreover, only few results are available on scheduling non-preemptive strict periodic tasks since their performance...

An important challenge in the domain of automotive control design is to provide a seamless flow for modelling conjointly with the behaviour, the temporal characteristics and the timing constraints of a system at different abstraction levels. In addition, this flow should provide analysis phases for validating the real-time behaviour of the...

An important challenge in the domain of automotive control design is to provide a seamless flow for modelling conjointly with the behaviour, the temporal characteristics and the timing constraints of a system at different abstraction levels. In addition, this flow should provide analysis phases for validating the real-time behaviour of the...

We consider the problem of scheduling tasks with strict periods combined with sporadic tasks. Both types of task have fixed priorities and are preemptive. For a task with a strict period, %it is necessary to prove that for any job of the task, the difference between its starting time and its release time must be identical for every job. Tasks...