MEC Proxy for Efficient Cache and Reliable Multi-CDN Video Distribution

Abstract

The massive consumption of media contents needs of network accelerators, which boost the media delivery and optimize the traffic volume crossing the network from servers to media players. Content Delivery Network (CDN) is the common network function to distribute in a cloud-manner the contents, enhancing media availability and distribution performance. However, high concurrency rates of media sessions can produce CDN performance degradations and outages that impact negatively the Quality of Experience (QoE). 5G Multi-access Edge Computing (MEC) architecture envisions a QoE-aware system at the network edge which performs analytics to enhance or boost media services. This paper provides a novel MEC proxy to expand and enforce caching infrastructures for efficient and reliable content distribution. First, the proxy caches contents on the network edge to reduce the Capital Expenditure (CAPEX) of the CDN for the OTT service provider. To this end, the proxy is able to identify recurrent requests. Second, the proxy shields from identified or predicted CDN malfunction. Here, the proxy switches the download sessions to an alternative CDN in order to ensure QoE rates, enabling a CDN dynamic selection based on live connectivity statistics. The proposed solution is evaluated by delivering Dynamic Adaptive Streaming over HTTP (MPEG-DASH) streams in a dense client cell while applying different caching strategies.