Repairable Fountain Coded Storage Systems for Multi-Tier Mobile Edge Caching Networks

Ye Wang, Shushi Gu, Lian Zhao, Ning Zhang, Wei Xiang, Qinyu Zhang

Abstract

Mobile edge caching (MEC) is an emerging paradigm where cloud services are extended to base stations (BSs) and devices to support a variety of applications and services in the Internet of Things (IoT). It expects to exploit storage resources deployed at different BSs and idle resources of end devices to efficiently cache and deliver contents. However, frequent mobility of devices with cached data can lead to data loss. Therefore, it becomes an important yet very challenging issue to cache the massive data with high data fault-tolerance and low cost. In this paper, we consider a multi-tier MEC network, where cloud-BS, edge-BSs and mobile devices collaboratively store and deliver contents to users. In addition, a framework based on repairable fountain codes (RFCs) with unequal repair locality (URL) for data caching, repairing and downloading, termed URL-RFC, is proposed to achieve efficient data caching in multi-tier heterogeneous MEC networks. Furthermore, a theoretical analysis and simulation results are provided to analyze and compare the energy costs of URL-RFC and existing solutions. It is demonstrated that the proposed URL-RFC scheme outperforms other redundant fault-tolerant schemes, which validates the potential value and feasibility of URL-RFC enabled storage in multi-tier MEC networks.
BibTeX
@article{wang2020repairable,
  author = {Ye Wang and Shushi Gu and Lian Zhao and Ning Zhang and Wei Xiang and Qinyu Zhang},
  title = {{Repairable Fountain Coded Storage Systems for Multi-Tier Mobile Edge Caching Networks}},
  journal = {IEEE Trans. Network Science \& Engineering},
  year = {2020},
  url = {https://openreview.net/forum?id=SQnO4P4w_w},
}

Loading related content...

Figure