Design of an efficient and resilient content distribution network

Abstract

The Internet has evolved to become a critical infrastructure for content delivery. Typical customers of a Content Distribution Network (CDN) are media and Internet advertisement companies, data centers, Internet Service Providers (ISPs), online music retailers, mobile operators, consumer electronics manufacturers, and other carrier companies. The main goal of the Content Distribution Network (CDN) is to replicate the content from the origin server to surrogate servers, scattered over the globe, in order to deliver the content to end users in a reliable and timely manner from nearby surrogate servers. The performance of content distribution network is analyzed to ensure that whether the user perceived Quality of Service (QoS) in content delivery is maintained. Quality of Service in content delivery is ensured by minimizing the impact of surrogate server failure as well as overcoming the overload problem that is a serious threat for busy sites serving popular contents. The main objective of this thesis is the application of the dominating set and its variants to construct the semantic overlay network of surrogate servers which in turn provides the logical infrastructure of the content distribution network. In this thesis, for designing an efficient and resilient content distribution network, the following algorithms are proposed to construct the logical infrastructure of the semantic overlay network. Dominating set formation algorithm; Equitable Dominating set formation algorithm; QoS aware Dominating set formation algorithm. The core of this thesis is constructing a QoS aware dominating set based semantic overlay network which satisfies the user perceived QoS constraints such as efficiency, controlled redundancy, and fault tolerant. The performance results of the proposed QoS aware Dominating set based semantic overlay network for content distribution is analyzed in terms of Size of CDN, Mean Response Time, Latency, Hit Ratio Percentage, Number of Completed Requests, Rejection rate and Mean CDN load.