Online Resource Allocation, Content Placement and Request Routing for Cost-Efficient Edge Caching in Cloud Radio Access Networks

Pu, Lingjun; Jiao, Lei; Chen, Xu; Wang, Lin; Xie, Qinyi; Xu, Jingdong

doi:10.1109/jsac.2018.2844624

Cited by 88 publications

(35 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It requires accurate resource orchestration regarding where, when, and how many service instances are deployed [31], [32]. Besides, the dependencies of services on each other [12], service migrations [13], load-balancing [14], and task scheduling [15] affect the costs of providers and also impact the user experience as well.…”

Section: Related Workmentioning

confidence: 99%

Service-Based Resilience via Shared Protection in Mission-Critical Embedded Networks

Ergenç

Rak

Fischer

2021

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

Mission-critical networks, which for example can be found in autonomous cars and avionics, are complex systems with a multitude of interconnected embedded nodes and various service demands. Their resilience against failures and attacks is a crucial property and has to be already considered in their design phase. In this paper, we introduce a novel approach for optimal joint service allocation and routing, leveraging virtualized embedded devices and shared backup capacity for the fault-tolerant design of mission-critical networks. This approach operates in phases utilizing multiple optimization models. Furthermore, we propose a new heuristic that ensures resource efficiency and faulttolerance against single node and link failures as prerequisite for resilience. Our experiments for different application scenarios indicate that our heuristic achieves results close to the optimum and provides 50% of capacity gain compared to a dedicated capacity protection scheme. Moreover, our heuristic ensures faulttolerance against at least 90% of all potential single node failures.

show abstract

Section: Related Workmentioning

confidence: 99%

Service-Based Resilience via Shared Protection in Mission-Critical Embedded Networks

Ergenç

Rak

Fischer

2021

IEEE Trans. Netw. Serv. Manage.

View full text Add to dashboard Cite

show abstract

“…where the updated dual variablesμ k andν k are obtained respectively by iteration (29) and (35). If we plugĉ f k instead of c f k into the marginalized Q-function in (21), then the solution for (P8) in current iteration k for a given file f can readily be found by solving arg min (w,a)∈X (rt,at−1)Q w,a rt,st + w(λ t +ν t σ f ) + a(ρ t +μ t σ f ). (37) Thus, it suffices to form a marginalized Q-function for each file and solve (37), which can be easily accomplished through exhaustive search over 8 possible cache-fetch decisions (w, a) ∈ X (r t , a t−1 ).…”

Section: Modified Online Solver Based On Q-learningmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

“…Targeting different objectives, optimization-based dynamic caching has been utilized in different approaches, see e.g., in [21]- [22]. To minimize content-access latency, energy, storage or bandwidth utilization costs, regularization and decomposition techniques have been used in [21]. Similar approaches are followed in [23] to relax a non-convex optimization problem to allocate limited caching memory across a network while accounting for the spatio-temporal content popularity together with the rented storage price fluctuations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Reinforcement Learning for Adaptive Caching With Dynamic Storage Pricing

Sadeghi

Sheikholeslami

Marqués

et al. 2019

IEEE J. Select. Areas Commun.

View full text Add to dashboard Cite

Small base stations (SBs) of fifth-generation (5G) cellular networks are envisioned to have storage devices to locally serve requests for reusable and popular contents by caching them at the edge of the network, close to the end users. The ultimate goal is to shift part of the predictable load on the backhaul links, from on-peak to off-peak periods, contributing to a better overall network performance and service experience. To enable the SBs with efficient fetch-cache decision-making schemes operating in dynamic settings, this paper introduces simple but flexible generic time-varying fetching and caching costs, which are then used to formulate a constrained minimization of the aggregate cost across files and time. Since caching decisions per time slot influence the content availability in future slots, the novel formulation for optimal fetch-cache decisions falls into the class of dynamic programming. Under this generic formulation, first by considering stationary distributions for the costs and file popularities, an efficient reinforcement learningbased solver known as value iteration algorithm can be used to solve the emerging optimization problem. Later, it is shown that practical limitations on cache capacity can be handled using a particular instance of the generic dynamic pricing formulation. Under this setting, to provide a light-weight online solver for the corresponding optimization, the well-known reinforcement learning algorithm, Q-learning, is employed to find optimal fetchcache decisions. Numerical tests corroborating the merits of the proposed approach wrap up the paper.

show abstract

“…The advantages of caching at the cloud level are: 1) BBUs have greater cache sizes, larger coverage areas and more easily to scalable compared with BSs or APs; 2) BBUs are closer to the users than the content servers in the core network; and 3) RRHs do not need to be modified, instead only focusing on the basic signal transmission functionalities [6], [7].…”

Section: Introductionmentioning

confidence: 99%

Joint Cache Allocation With Incentive and User Association in Cloud Radio Access Networks Using Hierarchical Game

Tran

et al. 2019

IEEE Access

View full text Add to dashboard Cite

In this paper, we consider a cloud radio access network-based system consisting of one network operator (NO) and several content providers (CPs). The NO owns a cloud cache and provides caching as a service for CPs, who provide contents to users. While the NO wishes to motivate CPs to rent its cache and maximize its profit, CPs want to optimize the service performance for users and their renting utilities. Due to the time separation between cache allocation and user association problems, we model the interactions between the NO and CPs as a hierarchical game, i.e., a cache renting scheme between the NO and CPs in the cache allocation problem and the willingness of CPs in the user association problem. In the cache allocation problem, we propose a contract theory-based incentive mechanism in which the NO designs and offers an optimal contract to various types of CPs. We then formulate the user association problem as a many-tomany matching game with externalities. To solve this matching game, we propose a matching algorithm that converges to a two-sided exchange stable matching with low complexity. The simulation results demonstrate that this proposed approach is beneficial to the NO's profit and incentivize the CP to rent the cache with truthful private information. In addition, the system performance of the proposed approach in terms of the total data rate-delay tradeoff outperforms than the benchmarks. INDEX TERMS Cloud RAN, caching as a service, cache allocation, contract theory, asymmetric information, matching game, externalities.

show abstract

Online Resource Allocation, Content Placement and Request Routing for Cost-Efficient Edge Caching in Cloud Radio Access Networks

Cited by 88 publications

References 32 publications

Service-Based Resilience via Shared Protection in Mission-Critical Embedded Networks

Service-Based Resilience via Shared Protection in Mission-Critical Embedded Networks

Reinforcement Learning for Adaptive Caching With Dynamic Storage Pricing

Joint Cache Allocation With Incentive and User Association in Cloud Radio Access Networks Using Hierarchical Game

Contact Info

Product

Resources

About