Secure Distributed Storage on Cloud-Edge Infrastructures

Abstract

Distributed storage systems place data in multiple cloud datacenters, leading to increased availability and flexibility. However, limitations are present when strict bandwidth and latency requirements are posed by the applications, as the data are stored into different, probably distant locations. The incorporation of edge resources in distributed storage services enables the placement of the data closer to their source, serving better the applications’ demands. Erasure coding offers a way to increase the availability and longevity of data during hosting. In our work, we develop mechanisms that perform resource allocation and store data at edge and cloud resources taking advantage of their different characteristics, while also exploiting the erasure coding technique. Initially, we provide a mixed integer linear programming formulation of the considered problem. As the search space can be vast and the execution time prohibitively large for real size problems, we also propose a heuristic approach which makes use of the rollout policy to efficiently trade-off performance with execution time. A set of simulation experiments is performed to showcase the validity of the proposed methods.