Generalization of LRU Cache Replacement Policy with Applications to Video Streaming

Abstract

Caching plays a crucial role in networking systems to reduce the load on the network and has become an ubiquitous functionality available at each router. One of the commonly used mechanisms, Least Recently Used (LRU), works well for identical file sizes. However, for asymmetric file sizes, the performance deteriorates. This paper proposes an adaptation to LRU strategy, called gLRU, where the file is sub-divided into equal-sized chunks. In this strategy, a chunk of the newly requested file is added in the cache, and a chunk of the least-recently-used file is removed from the cache. Even though approximate analysis for the hit rate has been studied for LRU, the analysis does not extend to gLRU since the metric of interest is no longer the hit rate as the cache has partial files. This paper provides a novel approximation analysis for this policy where the cache may have partial file contents. The approximation approach is validated by simulations. Further, gLRU outperforms LRU strategy for Zipf file popularity distribution and censored Pareto file size distribution for the file download times. Video streaming applications can further use the partial cache contents to help the stall durations significantly, and the numerical results indicate significant improvements (29%) in stall durations using the gLRU strategy as compared to the LRU strategy.

Modeling and Optimization of Latency in Erasure-coded Storage Systems

“…Efficient caching can help reduce both mean and tail latency. Different caching mechanisms based on Least-Resenly-Used strategy and its adaptations have been studied (Al-Abbasi and Aggarwal, 2020;Friedlander and Aggarwal, 2019). Erasure-coded mechanisms of caching have also been explored (Aggarwal et al, 2017a).…”

“…In the vCDN environment, we consider optimized caching at the CDN servers and LRU based caching at the edge router. However, with different file sizes, different caching algorithms have been studied (Berger et al, 2017;Halalai et al, 2017;Friedlander and Aggarwal, 2019). Considering efficient caching strategies for video streaming is an important problem for the future.…”

Joint Radio Resource Allocation and Cooperative Caching in PD-NOMA-Based HetNets

“…The size of content c, denoted by s c , is modeled by Log-normal distribution with the mean value µ c and variance σ 2 c . The Zipf distribution is a well-known model for content's popularity [7], [39]- [41] which is given by…”

Secure and Energy-Efficient Transmissions in Cache-Enabled Heterogeneous Cellular Networks: Performance Analysis and Optimization

“…2) SOP: Perfect secrecy is violated if an arbitrary partition of the requested file is intercepted by the eavesdroppers. Then, the SOP can be defined as the complement of the probability that log 2 1 + γ s k ,ej ≤ R e holds for all k ∈ {1, · · · , K}, where γ s k ,ej is the SNR of the wiretap channel from the k-th SBS to the j-th eavesdropper given in (5). Hence, we have > ω d .…”

“…Against this background, wireless caching techniques, via which popular content can be pre-stored at the edge of a wireless network before being requested by users, have emerged as a promising approach for relieving the backhaul bottleneck. Recently, wireless caching has been substantially proven to be highly effective in aspects of alleviating the enormous backhaul demand [1], [2], reducing the file service latency [3]- [5], increasing the successful delivery probability [6], and improving the energy efficiency [7], [8], etc.…”