Towards statistical estimation of worst case inter-core communications

Abstract : Statistical approaches have been recently received increased effort from the real-time community. While taking into account that worst case scenarios are rare events, the statistical approaches may decrease considerable the pessimism of a real-time analysis. In this paper we present first results on the statistical estimation of inter-core communications on a multicore platform. Our preliminary results are based on the utilization of the Extreme Value Theory and they are promising as they allow to detect insufficient observations of the communications.
Type de document :
Poster
JRWRTC 2017 - 11th Junior Researcher Workshop on Real-Time Computing, Oct 2017, Grenoble, France. 〈http://www.rtns17.org/#page=juniorworkshop〉
Liste complète des métadonnées

https://hal.inria.fr/hal-01666243
Contributeur : Liliana Cucu <>
Soumis le : lundi 18 décembre 2017 - 10:50:31
Dernière modification le : mercredi 20 décembre 2017 - 15:51:06

Identifiants

  • HAL Id : hal-01666243, version 1

Collections

Citation

Anselme Revuz, Liliana Cucu-Grosjean. Towards statistical estimation of worst case inter-core communications. JRWRTC 2017 - 11th Junior Researcher Workshop on Real-Time Computing, Oct 2017, Grenoble, France. 〈http://www.rtns17.org/#page=juniorworkshop〉. 〈hal-01666243〉

Partager

Métriques

Consultations de la notice

97