A data replica placement strategy for IoT workflows in collaborative edge and cloud environments

The convergence of edge and cloud computing shares their strengths, such as unlimited shared storage and computing resources from cloud, low-latency data preprocessing of edge computing. The collaboration of the two computing paradigms can provide a real-time and cost-effective way to deploy Internet of Things (loT) workflows among cooperative user groups. Since huge amounts of datasets continuously generated from user devices, how to place them to reduce the data access costs while meeting the deadline constraint is a critical issue. This paper proposed a novel data replica placement strategy for coordinated processing data-intensive loT workflows in collaborative edge and cloud computing environment. Firstly, data replica placement can be modelled as a 0-1 integer programming problem to consider the overall data dependency, data reliability and user cooperation. And then, the ITO algorithm, a variant of intelligent swarm optimization, is presented to address this model. The experimental results show that the proposed method outperforms these compared algorithms. It can not only find a higher quality solution of data replica placement, but also need a lower computing budget compared with these traditional algorithms. (C) 2018 Elsevier B.V. All rights reserved.