Data-intensive computing application based on workflow accounts for a large share in cloud computing. When processing data being stored in more than one data center. How to get data efficiently plays a big part in improving quality of service(QoS) and process execution efficiency. In this paper, a model is presented to describe the data intensive application. By measuring matched load of data nodes, a domain-based duplication strategy is also declared out. At last,the simulation results show that this strategy can improve the data acquisition efficiency markedly.