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

Abstract

In this paper, we consider the problem of allocating cache resources among multiple content providers. The cache can be partitioned into slices and each partition can be dedicated to a particular content provider, or shared among a number of them. It is assumed that each partition employs the LRU policy for managing content. We propose utility-driven partitioning, where we associate with each content provider a utility that is a function of the hit rate observed by the content provider. We consider two scenarios: i) content providers serve disjoint sets of files, ii) there is some overlap in the content served by multiple content providers. In the first case, we prove that cache partitioning outperforms cache sharing as cache size and numbers of contents served by providers go to infinity. In the second case, It can be beneficial to have separate partitions for overlapped content. In the case of two providers it is usually always benefical to allocate a cache partition to serve all overlapped content and separate partitions to serve the non-overlapped contents of both providers. We establish conditions when this is true asymptotically but also present an example where it is not true asymptotically. We develop online algorithms that dynamically adjust partition sizes in order to maximize the overall utility and prove that they converge to optimal solutions, and through numerical evaluations we show they are effective.

Cache Subsidies for an Optimal Memory for Bandwidth Tradeoff in the Access Network

“…When storage capacity or network bandwidth is limited, it is necessary to partition the resource between multiple CPs. Optimal partitioning of a limited capacity cache was considered by Dehghan et al [34] and Araldo et al [35]. Both papers propose iterative schemes where partitions are adjusted by the network operator based on observed per-CP performance.…”

On Resource Pooling and Separation for LRU Caching

“…Web and network caching is closely related to this study with a large body of dedicated works; see the surveys [82], [69] and the references therein. Recently a utility optimization approach [30], [35] based on the Che approximation [29], [18] has been used to study cache sharing and partitioning. It has concluded that under certain settings the optimal resource separation is better than pooling.…”

“…An optimal separation scheme has been proposed in [35] to serve classes A, B, D in three isolated parts of the whole cache space. Since the three classes do not have overlapped data items, Theorem 3 implies that the optimal separation is asymptotically equivalent to pooling.…”

“…Our analysis shows that for large cache spaces it is beneficial to jointly serve multiple flows if their data item sizes and popularity distributions are similar and their arrival rates do not differ significantly. Although the optimal static resource separation scheme has been shown to always theoretically achieve the best performance under certain assumptions [35], in practice the number of separate clusters deployed in service, e.g., Memcached clusters, is relatively small [9]. This may be partially attributed to the self-organizing behavior of LRU for resource pooling.…”

Radio Resource Sharing and Edge Caching with Latency Constraint for Local 5G Operator: Geometric Programming Meets Stackelberg Game

“…Apart from the efficient resource utilization, achieving low latency for xMBB with partial/full virtual reality is also a critical challenge for the MNOs. To tackle this challenge, the concept of proactive caching was introduced [16]- [19] in which popular contents are stored at the edge/radio access network (RAN), e.g., cacheenabled BSs to reduce the wireless access delay. In [15], the virtualization technique in the downlink transmission of limited fronthaul capacity cloud-radio access networks was considered.…”