SLA Establishment with Guaranteed QoS in the Interdomain Network: A Stock Model

📝 Abstract

The new model that we present in this paper is introduced in the context of guaranteed QoS and resources management in the inter-domain routing framework. This model, called the stock model, is based on a reverse cascade approach and is applied in a distributed context. So transit providers have to learn the right capacities to buy and to stock and, therefore learning theory is applied through an iterative process. We show that transit providers manage to learn how to strategically choose their capacities on each route in order to maximize their benefits, despite the very incomplete information. Finally, we provide and analyse some simulation results given by the application of the model in a simple case where the model quickly converges to a stable state.

💡 Analysis

The new model that we present in this paper is introduced in the context of guaranteed QoS and resources management in the inter-domain routing framework. This model, called the stock model, is based on a reverse cascade approach and is applied in a distributed context. So transit providers have to learn the right capacities to buy and to stock and, therefore learning theory is applied through an iterative process. We show that transit providers manage to learn how to strategically choose their capacities on each route in order to maximize their benefits, despite the very incomplete information. Finally, we provide and analyse some simulation results given by the application of the model in a simple case where the model quickly converges to a stable state.

📄 Content

International Journal of Computer Networks & Communications (IJCNC) Vol.3, No.4, July 2011 DOI : 10.5121/ijcnc.2011.3413 188

SLA ESTABLISHMENT WITH GUARANTEED QOS IN THE INTERDOMAIN NETWORK: A STOCK MODEL ∗ Dominique Barth1, Boubkeur Boudaoud1 and Thierry Mautor1 1 PRiSM, University of Versailles-Saint-Quentin, 45 Av. des Etats-Unis 78000 France dominique.barth@prism.uvsq.fr
boubkeur.boudaoud@prism.uvsq.fr thierry.mautor@prism.uvsq.fr

ABSTRACT The new model that we present in this paper is introduced in the context of guaranteed QoS and resources management in the inter-domain routing framework. This model, called the stock model, is based on a reverse cascade approach and is applied in a distributed context. So transit providers have to learn the right capacities to buy and to stock and, therefore learning theory is applied through an iterative process. We show that transit providers manage to learn how to strategically choose their capacities on each route in order to maximize their benefits, despite the very incomplete information. Finally, we provide and analyse some simulation results given by the application of the model in a simple case where the model quickly converges to a stable state. KEYWORDS Learning theory, distributed algorithmic, QoS and Resource Management, inter-domain routing.

1. INTRODUCTION In the inter-domain network routing, the service Level Agreements (SLAs) correspond to the contracts between Internet Service Providers (ISPs) and their users for services that ISPs provide to their customers. For the Benefit of the ISPs and their users in the inter-domain, the concept of SLA has been introduced. These SLAs are characterized by the informations of guaranteed QoS and accounting of services. They allow to identify and define customer needs. Both QoS-guaranteed services and the accounting cost are critical to the SLA as far as Internet services are concerned [10]. A service contract or SLA in the inter-domain, is a bipartite agreement between customer and service provider. According to [4] the negotiation between both parties (customer and provider) reflects SLS (Service Level Specification) which contains a set of technical parameters that define the level of service provided by a domain. Provisioning is one of the phases that SLA contains in its life cycle [3] and [4], this phase consists in reserving resources and activating the service. In this paper we show how to implement the management of resources and inter-domain QoS guaranteed services, while taking into account the economic constraints of the various operators in the inter-domain network. For this implementation, we have used a new model that we have called stock model. In order to reduce the complexity of QoS indicators, we express them by only two parameters corresponding to the equivalent bandwidth and transmission delay of traffic. In our work, we consider that the customers buy the routes with guaranteed QoS from their transit providers. The goal of purchasing a route by an operator is to allow its customers to reach a desired service in the destination of this route. The purchase of a route is based on the reservation of the guaranteed temporal capacity on this route, the transmission delay of traffic

∗ This work is supported by the FP7 european project “ETICS: Economics and Technologies for Inter-Carrier Services” International Journal of Computer Networks & Communications (IJCNC) Vol.3, No.4, July 2011 189

must be also guaranteed on this route. This temporal capacity is available over a precise period of time (eg, a unit of time is one hour), and the transmission delay on this route is defined as the sum of transmission delays between each pair of adjacent nodes from the customer until the destination. For purchasing a route to a given destination, we consider that each domain in the network negotiates with its neighbor (transit provider) the capacity and its availability on this route. In this case, the customer asks to its provider for a minimum capacity1 and a maximum capacity2, it also asks for a period of availability3 of the capacity asked on this route. This customer must also choose a start time4 of a period asked. Each contract established between one customer and its provider for a route to a given destination has the following characteristics: the capacity determined by the provider between the minimum capacity and the maximum capacity, the period of availability and its start time chosen by the customer, the transmission delay of traffic and the price of this route. [2][5][7] and [8] have addressed the comparison with usage dependent pricing. After the establishment of this contract, the customer may negotiate with its transit provider the decrease or the increase of the

This content is AI-processed based on ArXiv data.