Guaranteed service level agreements across multiple ISP networks

Due to the autonomous nature of ISPs, the service level agreement (SLA) offering is currently confined to within a single provider network. In this work, we examine some methods of extending the SLA offering across ISP boundaries. We introduce three policies to coordinate the end-to-end performance guarantee in multiple ISP networks: the least-effort, the most-effort, and the equal-distribution policies. These policies refer to different manners in which the service-level constraints are distributed among all transit networks. We study the impacts of these policies on the overall ISP community when SLA is required. We evaluate the effectiveness of these policies in terms of both the network performance and the ISP's monetary profit. The results show that the policy choice depends on the network load, as well as the ISP cost structure.

I. Introduction

A service level agreement (SLA) is a contract between an Internet service provider (ISP) and its customer. SLAs obligate service providers to maintain a certain grade or level of service. Service providers are keen on offering SLAs because SLAs permit differential treatment of the customer traffic. This differential treatment can yield economic benefit to the service providers by, for example, not having to allocate a protection path to all connections. This way the providers can make the best use of their resources given their SLA con-straints. On the other hand, the customer wants SLA guarantee because they can ensure the rigid level of performance they pay for, and be compensated for the lack thereof. SLAs also provide an incentive for the customer to subscribe for services since they can choose the level that suits their need, and not have to pay premium for unnecessary features.