Set-Covering Theory-Based Data Placement Cost Optimization for Online Social Networks

摘要

As the web evolves and rapid development of the intelligence devices, both the number of users and online social network data are increasing rapidly every day. It is a fundamental issue to consider how to find an appropriate data placement to reduce the cost of data storage while meeting users' latency requirements. Therefore, our objective is to optimize the total cost of data storage while guaranteeing all users' latency requirements. We transform the data placement in cloud of online social networks to the minimum set covering problem through the model of latency constrained matrix. Based on this model, this paper proposes a heuristic data placement algorithm called LDSA to achieve our goal. Experiments demonstrate that the proposed algorithm can significantly reduce the cost of data storage while guaranteeing all users' latency requirements and has an excellent time performance in comparison with other representative placement strategies.