Optimal content placement for adaptive bit-rate streaming in cache networks

In the cache networks, bringing the contents closer to the end users by caching them at the edge nodes helps to reduce the transit traffic and to increase the quality-of-experience of the end users. Recently, adaptive bit-rate (ABR) streaming becomes a major technique to deliver the video contents to the end users in the Internet. With ABR, each video content is stored with several levels of representations corresponding to multiple bit-rates. A higher bit-rate representation yields a higher user satisfaction, but consumes more storage. Given the content popularity distribution, this paper proposes an optimal content placement model maximizing the sum of user satisfaction of all contents. The content placement problem is usually a large-scale optimization problem due to the large number of content items in the network. With mild assumption on the probability density function of the content popularity distribution, we reduce the large-scale problem to an equivalent small-scale convex problem in which the number of variables is only the number of levels of representations.