End-to-end delay bound calculation in generalized processor sharing networks

Abstract

In this paper we propose a solution to the problem of calculating deterministic end-to-end delay bounds in arbitrary topology networks, which employ the Generalized Processor Sharing (GPS) discipline at each node. We give a general framework under network calculus [ I ] by extending the service curve model of GPSand constructing the end-to-end service curve for each session, from which exact values of delay and backlog bounds can be computed. Network internal traffic characterization and traffic shaping are applied to support local pe@ormance bound computation. Sessions are allowed to be arbitraiy weighted and virtual feedback is handled by the iterative approach of [2]. Numerical examples are shown to illustrate our algorithmic approach for per-session end-to-end delay bound computation.

Modelagem e determinação de parametros de desempenho de redes de comunicações atraves da algebra de dioides

“…Por exemplo, Goyal et al (1997) determinam medidas de atraso fim-a-fim, quando o algoritmo de escalonamento é do tipo Guaranteed Rate (GR - Goyal et al, 1997). Barta et al (2001) determinam atraso fim-a-fim em redes que empregam a disciplina GPS (Generalized Processor Sharing -Parekh e Gallager, 1993) em cada nó. Outros exemplos de trabalhos nesta linha são:…”