A Multi-Agent Algorithm to Improve Content Management in CDN Networks

An effective solution to delivery static contents are the Content Delivery Networks (CDNs). However, when the network size increases, they show limits and weaknesses related to their size, dynamic nature, and due to the centralized/heirarchical algorithms used for their management. Decentralized algorithms and protocols can be usefully employed to improve their efficiency. A bio-inspired algorithm that improves the performance of CDNs by means of a logical organization of contents is presented in this paper. Self-organizing ant-inspired agents move and organize the metadata describing the content among the CDN servers, which are interconnected in a peer-to-peer fashion, so as to improve discovery operations. Experimental results confirm the effectiveness of the adopted approach.