Distributed and Decentralized Edge Caching in 5G Networks Using Non-Volatile Memory Systems

Edge caching is an effective way to reduce congestion and latency in 5G networks. Non-volatile memory (NVM) devices are developing fast, with the potential of fast access, and higher endurance versus traditional storage devices, to further boost mobile data offloading efficiency in 5G networks. This...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings of the International Conference on Distributed Computing Systems pp. 425 - 435
Main Authors: Zeng, Yiming, Huang, Yaodong, Liu, Zhenhua, Liu, Ji
Format: Conference Proceeding
Language:English
Published: IEEE 01.07.2022
Subjects:
5G mobile communication
5G networks
Delays
distributed and decentralized algorithms
Distributed databases
edge caching
non-volatile memory
Nonvolatile memory
Numerical simulation
Performance evaluation
Routing
ISSN:2575-8411
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Edge caching is an effective way to reduce congestion and latency in 5G networks. Non-volatile memory (NVM) devices are developing fast, with the potential of fast access, and higher endurance versus traditional storage devices, to further boost mobile data offloading efficiency in 5G networks. This paper studies how to effectively use the two-layer storage system (NVM-enhanced) in 5G edge caching. We first model an edge caching optimization problem with NVM storage devices included and develop a parallel distributed algorithm with guaranteed convergence in joint caching and routing decisions. A fully decentralized algorithm for scenarios without any coordination is further developed which also guarantees the convergence. Real-world trace-driven simulations and experiments over a small-scale system demonstrate that NVM significantly boosts the performance of edge caching and the proposed algorithms outperform the existing ones.
ISSN:2575-8411
DOI:10.1109/ICDCS54860.2022.00048