Sharing Cache Resources Among Content Providers: A Utility-based Approach

Sharing Cache Resources among Content Providers: A Utility-Based Approach

Abstract

In this paper, we consider the allocation problem Cache resources in multiple content providers. Cache can divide it into pieces and you can dedicate every partition shared to a specific content provider or many them. Each partition is assumed to be working recently the approach used for content management. We propose a utility-drive partition, where we associate each content function of the observed hit rate Content provider. We consider two scenarios: 1) content providers provide a set of files and 2) there is some overlap Content provided by multiple content providers. In the first case, we have demonstrated that cache partitioning outweighs cache sharing. Cache size and many other things providers have to offer infinity. In the latter case, separation is beneficial for partitions of overlapping content. In the case of two providers, assigning a cache partition is usually always advantageous providing all overlapping content and separate partitions to serve the overlapping of the two providers. We have established circumstances when it's ineffectively true but this is not ineffectively true, for example. We develop online algorithms that dynamically adjust partition sizes, increase the total utility and prove that they converge. We show through correct solutions and numerical evaluations they are effective.