A Storage Resource Collaboration Model Among Edge Nodes in Edge Federation Service

The edge computing service mode has entered the edge federation service stage, where comprehensive resources are supported to collaborate between vertical cloud-edge and horizontal edge nodes. The massive data processing poses a severe challenge to the traditional storage mechanism. In this paper, we propose an efficient storage resource collaboration model that leverages the selection of storage collaborative edge nodes and the allocation of storage resources among collaborative nodes, which can provide efficient and balanced storage services solutions. Specifically, selecting storage collaborative edge nodes is a multi-objective integer linear programming problem. We transform it into a single objective 0-1 integer linear programming problem and use the dynamic programming method to obtain the optimal solution. In addition, we condense the storage resource allocation problem into a mixed-linear integer programming problem and use the greedy auction algorithm to obtain an approximate optimal solution, which significantly improves efficiency and performance. We evaluate the performance of the proposed two mechanisms by performing extensive experiments. The experimental results show that they are efficient solutions to solve the problem of the storage resource collaboration model among edge nodes in edge federation service.