High-speed packet networks will begin to support services that need Quality-of-Service (QoS) guarantees. Guaranteeing QoS typically translates to reserving resources for the duration of a call. We propose a statedependent routing scheme that builds on any base stateindependent routing scheme, by routing flows which are blocked on their primary paths (as selected by the state-independent scheme) onto alternate paths in a manner that is guaranteed---under certain Poisson assumptions---to improve on the performance of the base state-independent scheme. Our scheme only requires each node to have state information of those links that are incident on it. Such a scheme is of value when either the base state-independent scheme is already in place and a complete overhaul of the routing algorithm is undesirable, or when the state (reserved flows) of a link changes fast enough that the timely update of state information is infeasible to all possible call-originators. The performance improvements ...

and its Working Groups. Note that other groups may also distribute working documents as Internet Drafts. Internet Drafts are draft documents valid for a maximum of six months. Internet Drafts may be updated, replaced, or obsoleted by other documents at any time. It is not appropriate to use Internet Drafts as reference material or to cite them other than as a "working draft " or "work in progress." To learn the current status of any Internet Draft, please check the 1id-abstracts.txt listing contained in the Internet Drafts Shadow Directories on ds.internic.net, nic.nordu.net, ftp.nisc.sri.com or munnari.oz.au. Distribution of this memo is unlimited. This Internet Draft expires on September, 2, 1998. QoS-based routing has been recognized as a missing piece in the evolution

Supporting guaranteed services in a large network requires solutions to many difficult problems. In this paper, we focus on the problem of on-demand routing of virtual circuits with quality of service guarantees. Specifically, we investigate the performance of the ATM Private NetworkNetwork Interface (PNNI) protocol in a large topology with structure reflecting the hierarchy and locality present in internetworks. PNNI uses several techniques to maintain scalability of the routing process; each technique has an associated penalty in performance (i.e., the ability of the protocol to find routes). The tension between scalability and performance is the subject of this paper. We examine the routing performance under a variety of aggregation and topology update conditions. The work described in this paper uses a large-scale simulator being developed as part of the Scalable Self-Organizing Simulation project. 1 Introduction 1.1 Problem Statement Supporting integrated and guaranteed network...

Abstract. This paper presents several decentralized learning algorithms for on-line intra-domain routing of bandwidth guaranteed paths in MPLS networks when there is no a-priori knowledge of traffic demand. The presented routing algorithms use only their locally observed events and update their routing policy using learning schemes. The employed learning algorithms are either learning automata or the multi-armed bandit algorithms. We investigate the asymptotic behavior of the proposed routing algorithms and prove the convergence of one of them to the user equilibrium. Discrete event simulation results show the merit of these algorithms in terms of increasing the network admissibility compared with shortest path routing. We investigate the performance degradation due to decentralized routing as opposed to centralized optimal routing policies in practical scenarios. The system optimal and the Nash bargaining solutions are two centralized benchmarks used in this study. We provide nonlinear programming formulations of these problems along with

Abstract-In this paper, we present an approximate model (with finite or infinite waiting room) for an integrated service system with three types of traffic: a first offered narrow-band traffic, an overflow narrow-band traffic and a wide-band traffic. A narrow-band call requires a single server, while the number of servers required to serve a wide-band call is N. The blocked narrow-band calls are lost while the blocked wide-band calls are delayed in a finite or infinite waiting room. Based on two assumptions with regard to the characteristics of the system, we resolve the system by decomposition. The corresponding improvements in numerical efficiency as well as in computational storage requirements are significant enough to enable use of the model within network optimization algorithms. The model provides a very good approximation for the system performance, that is the blocking probabilities of the two narrowband traffics, the loss probability (in the case of finite waiting room), the probability of nonwaiting and the average waiting time of wide-band traffic. I.