A user interest-aware content replica optimized placement algorithm for CDN networks

The replica placement algorithm plays a key role in improving the distribution efficiency of content delivery networks( CDNs). However, most existing algorithms consider only increasing the ISPs´ benefits, and do not consider matching the user´s interesting content subjects and improving its QOE for CDN services. Therefore, this paper proposes a user interestaware content replica optimized placement algorithm (UIARP). Firstly, the user´s interest is extracted from its content access logs by clustering algorithms. Then replicas are placed according to the descending order of the collective interest degree, which satisfies the constraints of responsive tolerance limit and storage capacity. Finally, the UIARP algorithm can achieve the match between placing replicas and user content demand through minimizing the average response time. This algorithm not only ensures that users get interested replicas quickly, but also improves the network performance. The simulation analyses from four aspects including the average response time, the request response matching degree, load balancing and the adjacent replica utilization rate, which verify the effectiveness of the proposed algorithm.