RTLib: A Library of Timed Automata for Modeling Real-Time Systems - Archive ouverte HAL Access content directly
Reports (Research Report) Year : 2016

RTLib: A Library of Timed Automata for Modeling Real-Time Systems

Lijun Shan
  • Function : Correspondent author
  • PersonId : 992959

Connectez-vous pour contacter l'auteur
Susanne Graf

Abstract

There exists a large variety of schedulability analysis tools based on different, often incomparable timing analysis models. This variety makes it difficult to choose the best fit for analyzing a given real-time system. We have developed RTLib, a library of timed automata templates that represent the semantics of a large variety of timing related concepts proposed by existing models and corresponding timing analysis tools. The key specificity of RTLib is that it is structured to be modular, such that (1) defining a new variant of a given concept requires only a localized change, and (2) non contradictory variants can be trivially combined. The extensibility of RTLib is demonstrated using five examples ranging from simple variants of the task activation model to the more complex mixed-criticality paradigm. RTLib provides the formal basis needed to compare the concepts offered by models of different timing analysis tools at the semantic level. This in turn will allow us to provide a syntactic mapping between the input of different tools. Our final goal is to help the research community to better evaluate analysis models and their underlying methods.
Fichier principal
Vignette du fichier
short.pdf (317.69 Ko) Télécharger le fichier
Origin Files produced by the author(s)
Loading...

Dates and versions

hal-01393888 , version 1 (08-11-2016)

Identifiers

  • HAL Id : hal-01393888 , version 1

Cite

Lijun Shan, Susanne Graf, Sophie Quinton. RTLib: A Library of Timed Automata for Modeling Real-Time Systems. [Research Report] Grenoble 1 UGA - Université Grenoble Alpe; INRIA Grenoble - Rhone-Alpes. 2016. ⟨hal-01393888⟩
424 View
559 Download

Share

Gmail Mastodon Facebook X LinkedIn More