2020Proceedings of the ACM on Measurement and Analysis of Computing SystemsRequires access

On Private Peering Agreements between Content and Access Providers

Xin Wang, T. B. Richard

Open publisher page 5 citations

Abstract

Driven by the rapid growth of content traffic and the demand for service quality, Internet content providers (CPs) have started to bypass transit providers and connect with access providers directly via private peering agreements. This peering relationship often raises disputes, e.g., Netflix vs. Comcast, and is not well understood. In this paper, we build a peering contract model and propose the concept of contractual equilibrium, based on which we study the formation and evolution of peering contracts. By using market data, we emulate the strategic peering behavior of providers and shed light on the understanding of private peering agreements. We reveal that the superiority and market dominance of providers primarily determine their peering strategies. We show that 1) superior providers tend to engage in peering more aggressively, and 2) non-dominant CPs' optimal peering strategies are negatively correlated due to market cannibalism, while the dominant CP often behaves oppositely. Our findings help explain phenomena such as why Netflix and Comcast signed the first peering contract, and reason whether private peering contracts will strengthen in future.

About this research paper

What this paper is about

Driven by the rapid growth of content traffic and the demand for service quality, Internet content providers (CPs) have started to bypass transit providers and connect with access providers directly via private peering agreements. This peering relationship often raises disputes, e.g., Netflix vs. Comcast, and is not well understood. In this paper, we build a peering contract model and propose the concept of contractual equilibrium, based on which we study the formation and evolution of peering contracts. By using market data, we emulate the strategic peering behavior of providers and shed light on the understanding of private peering agreements. We reveal that the superiority and market dominance of providers primarily determine their peering strategies. We show that 1) superior providers tend to engage in peering more aggressively, and 2) non-dominant CPs' optimal peering strategies are negatively correlated due to market cannibalism, while the dominant CP often behaves oppositely. Our findings help explain phenomena such as why Netflix and Comcast signed the first peering contract, and reason whether private peering contracts will strengthen in future.

Why it matters

OpenAlex reports 5 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Driven by the rapid growth of content traffic and the demand for service quality, Internet content providers (CPs) have started to bypass transit providers and connect with access providers directly via private peering agreements. This peering relationship often raises disputes, e.g., Netflix vs. Comcast, and is not well understood. In this paper, we build a peering contract model and propose the concept of contractual equilibrium, based on which we study the formation and evolution of peering contracts. By using market data, we emulate the strategic peering behavior of providers and shed light on the understanding of private peering agreements. We reveal that the superiority and market dominance of providers primarily determine their peering strategies. We show that 1) superior providers tend to engage in peering more aggressively, and 2) non-dominant CPs' optimal peering strategies are negatively correlated due to market cannibalism, while the dominant CP often behaves oppositely. Our findings help explain phenomena such as why Netflix and Comcast signed the first peering contract, and reason whether private peering contracts will strengthen in future.

Key concepts: Peering, Tier 2 network, Business, Tying, Tier 1 network, Dominance (genetics), Service provider, Internet service provider

Related papers

Back to paper searchBrowse research topicsOriginal source
On Private Peering Agreements between Content and Access Providers — Research Paper | ScholarLens