DR-Cache: Distributed Resilient Caching with Latency Guarantees

Li, Jian; Phan, Truong Khoa; Chai, Wei Koong; Tuncer, Daphné; Pavlou, George; Griffin, David; Rio, Miguel

doi:10.1109/infocom.2018.8486316

Cited by 46 publications

(33 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Related Workmentioning

confidence: 99%

“…During recent years, there has been great interests in proactive cooperative content caching [8]- [14]. In [8]- [10], the authors designed polices to minimize the aggregate network cost due to content transfers across the network. In [11], a caching policy was defined to minimize the content downloading delay by taking into account cache deployment costs in the network.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Online Spatiotemporal Popularity Learning via Variational Bayes for Cooperative Caching

Mehrizi

Chatterjee

Chatzinotas

et al. 2020

IEEE Trans. Commun.

View full text Add to dashboard Cite

Herein, we focus on an end-to-end design of a proactive cooperative caching strategy for a multi-cell network. The design is challenging as it involves two interrelated problems: the ability to predict future content popularity and to meet network operation characteristics. To this end, we first formulate a cooperative content caching in order to optimize the aggregated network cost for delivering contents to users. An efficient proactive caching policy requires an accurate prediction of timevarying content popularity. Content popularity has temporal and spatial dependencies and therefore, we develop a probabilistic dynamical model for content popularity prediction by exploiting its spatiotemporal correlations. To achieve an accurate tracking and prediction of content popularity evolution, the proposed dynamical model is non-linear and incorporates non-Gaussian distributions. We use Variational Bayes (VB) approach for estimating the model parameters. The VB provides mathematical tractability. We then develop an online VB method that works with streaming data where content request arrives sequentially. Using extensive simulations study on a real-world dataset, we show that our online VB based dynamical model provides improved performance compared to conventional content caching policies.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Online Spatiotemporal Popularity Learning via Variational Bayes for Cooperative Caching

Mehrizi

Chatterjee

Chatzinotas

et al. 2020

IEEE Trans. Commun.

View full text Add to dashboard Cite

show abstract

“…They proposed an efficient decentralized service placement algorithm based on graph coloring on the small cell network. Li et al [30] studied optimal content allocation over a resilient caching network, where each cache may fail under some situations. They proposed a centralized algorithm which achieves a (1 − 1 e )−approximation guarantee.…”

Section: Related Workmentioning

confidence: 99%

A New Framework for Mobile Edge Caching by Proposing Flexible User in Heterogeneous Cellular Networks

2020

View full text Add to dashboard Cite

The bursting increase in requesting wireless data has caused several issues in network peaktraffic duration. This negatively results in significant data delivery delay imposed on users that can eventually impact the network's quality of service and users' quality of experience. In this research, regarding mobile edge caching as a potential solution to decrease such delay, we propose a new framework in which we introduce the concept of the flexible user where he requests for a set of multiple files from the library with a unique feature, e.g., 5 movies within comedy genre from the library in the peak-traffic duration. The satisfactory criterion for the flexible user is to receive any of the files within the requested set. This definition of the flexible user indicates a new concept which captures interesting scenarios. In order to model this concept, we generalize the conventional Zipf distribution to a multivariate one as the modeling method for popular data. We formulate the problem of finding the optimal cache data placement, which minimizes the average total delivery delay in the network while satisfying the helpers' cache size constraints. To this end, we derive the average delivery delay per user as well as the average total delivery delay in the network, according to the new generalized Zipf distribution. Finding the optimal solution is proved to be NP-Hard. We leverage on the problem property to propose an efficient approximation method, called greedy algorithm, which performs within a constant factor as good as the optimal solution. Afterwards, we propose an algorithm called speedy-greedy to significantly reduce the computational complexity of the greedy algorithm while achieving the same performance. Simulation results indicate that our proposed framework significantly decreases the average total delivery delay of the system model that can help the network maintain its quality of service in network peak-traffic duration. INDEX TERMS Mobile edge caching, data delivery and management, delivery delay, femto-caching.

show abstract

“…In [12] and [13], the dynamic evolution of content popularity was investigated by using differential equations and Markov chains, respectively. In [14], a distributed gradient ascent algorithm was proposed to resiliently maximize the latency reduction in an Internet cache topology. In [15], the authors considered a fairness problem in a data-sharing network, and proposed a distributed algorithm that can continuously select nodes to cache contents adaptive to newly generated data.…”

Section: Introductionmentioning

confidence: 99%

A Mean Field Game-Based Distributed Edge Caching in Fog Radio Access Networks

Jiang

Bennis

et al. 2020

IEEE Trans. Commun.

View full text Add to dashboard Cite

In this paper, the edge caching optimization problem in fog radio access networks (F-RANs) is investigated. Taking into account time-variant user requests and ultra-dense deployment of fog access points (F-APs), we propose a distributed edge caching scheme to jointly minimize the request service delay and fronthaul traffic load. Considering the interactive relationship among F-APs, we model the optimization problem as a stochastic differential game (SDG) which captures the dynamics of F-AP states. To address both the intractability problem of the SDG and the caching capacity constraint, we propose to solve the optimization problem in a distributive manner. Firstly, a mean field game (MFG) is converted from the original SDG by exploiting the ultra-dense property of F-RANs, and the states of all F-APs are characterized by a mean field distribution. Then, an iterative algorithm is developed that enables each F-AP to obtain the mean field equilibrium and caching control without extra information exchange with other F-APs. Secondly, a fractional knapsack problem is formulated based on the mean field equilibrium, and a greedy algorithm is developed that enables each F-AP to obtain the final caching policy subject to the caching capacity constraint. Simulation results show that the proposed scheme outperforms the baselines. Index Terms-Fog radio access networks, distributed edge caching, mean field game, fractional knapsack problem, request service delay, fronthaul traffic load.

show abstract

DR-Cache: Distributed Resilient Caching with Latency Guarantees

Cited by 46 publications

References 28 publications

Online Spatiotemporal Popularity Learning via Variational Bayes for Cooperative Caching

Online Spatiotemporal Popularity Learning via Variational Bayes for Cooperative Caching

A New Framework for Mobile Edge Caching by Proposing Flexible User in Heterogeneous Cellular Networks

A Mean Field Game-Based Distributed Edge Caching in Fog Radio Access Networks

Contact Info

Product

Resources

About