Skip to Main content Skip to Navigation
Conference papers

FP/FIFO scheduling: deterministic versus probabilistic QoS guarantees and p-schedulability

Leila Azouz Saidane 1 Skander Azzaz 1 Steven Martin 2 Pascale Minet 3
1 CRISTAL - Centre de Recherche Réseau Image SysTème Architecture et MuLtimédia
2 LRI - Laboratoire de Recherche en Informatique
3 HIPERCOM - High performance communication
Inria Paris-Rocquencourt, UP11 - Université Paris-Sud - Paris 11, Inria Saclay - Ile de France, X - École polytechnique, CNRS - Centre National de la Recherche Scientifique : UMR
Abstract : In this paper, we focus on applications having quantitative QoS (Quality of Service) requirements on their end-to-end response time. Two types of quantitative QoS garantees can be delivered by a network, deterministic and probabilistic. The deterministic approach is based on a worst case analysis. The probabilistic approach uses a mathematical model to obtain the probability that the response time exceeds a given value. We assume that flows are scheduled according to non-preemptive FP/FIFO. The packet with the highest fixed priority is scheduled first. If two packets share the same fixed priority, the packet arrived first on the node considered is scheduled first. We compare deterministic and probabilistic QoS guarantees and introduce the concept of p-schedulabilty: the QoS requested by all flows is met with probability p.
Document type :
Conference papers
Complete list of metadata
https://hal.inria.fr/inria-00128777
Contributor : Pascale Minet <>
Submitted on : Friday, February 2, 2007 - 4:10:48 PM
Last modification on : Wednesday, October 28, 2020 - 12:48:02 PM

Identifiers

  • HAL Id : inria-00128777, version 1

Collections

INRIA | CNRS | UMR8623 | X

Citation

Leila Azouz Saidane, Skander Azzaz, Steven Martin, Pascale Minet. FP/FIFO scheduling: deterministic versus probabilistic QoS guarantees and p-schedulability. IEEE International Conference on Communications - ICC 2007, Jun 2007, Glasgow, Scotland / Great Britain. ⟨inria-00128777⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

1670

 

Contact                    Legal Notice                    Personal Data