Skip to Main content Skip to Navigation
Conference papers

A QoS Monitoring System for Dataflow Programs

Manuel Selva 1, * Lionel Morel 1, * Kevin Marquet 1 Stéphane Frénot 2
* Corresponding author
1 SOCRATE - Software and Cognitive radio for telecommunications
Inria Grenoble - Rhône-Alpes, CITI - CITI Centre of Innovation in Telecommunications and Integration of services
2 DICE - Data on the Internet at the Core of the Economy
Inria Grenoble - Rhône-Alpes, CITI - CITI Centre of Innovation in Telecommunications and Integration of services
Abstract : With the generalization of multi-core processors, dataflow programming is regaining a strong interest, especially in the context of compute intensive multimedia applications such as video decoding. How- ever, most studies focus on static approaches to the compilation and placement problems. We advocate for dynamic adaptation of dataflow applications. In this paper, we build the first step towards this goal, namely a monitoring mechanism for observing quality-of-service properties of programs at run- time. We propose a language extension for expressing simple QoS properties over dataflow programs together with a run-time mechanism for the observation of events meaningful to the QoS establishment. We show the limited impact of such mechanisms on the application overall performances.
Complete list of metadatas

Cited literature [4 references]  Display  Hide  Download

https://hal.inria.fr/hal-00780976
Contributor : Manuel Selva <>
Submitted on : Friday, January 25, 2013 - 10:51:42 AM
Last modification on : Wednesday, November 20, 2019 - 7:32:17 AM
Document(s) archivé(s) le : Friday, April 26, 2013 - 3:55:17 AM

File

cfse.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00780976, version 1

Collections

Citation

Manuel Selva, Lionel Morel, Kevin Marquet, Stéphane Frénot. A QoS Monitoring System for Dataflow Programs. ComPAS: Conférence en Parallélisme, Architecture et Système, Inria, Jan 2013, Grenoble, France. ⟨hal-00780976⟩

Share

Metrics

Record views

692

Files downloads

382