Cloud computing is an emerging pattern that provides computing, communication and storage resources as a service over a network. In existing system, data outsourced in a cloud is unsafe due to the eaves dropping and hacking process. And it allows minimizing the security network delays in cloud computing. In this paper to study data replication in cloud computing data centers. Unlike another approaches available in the literature, consider both security and privacy preserving in the cloud computing. To overcome the above problem we use DROPS methodology. The data encrypted using AES (Advanced Encryption Standard Algorithm). In this process, the common data are divided into multiple nodes also replicate the fragmented data over the cloud nodes. Each data is stored in a different node in fragments individual locations. We ensure a controlled replication of the file fragments, here each of the fragments is replicated only once for the purpose of improved security. The results of the simulations revealed that the simultaneous focus on the security and performance, resulted in improved security level of data accompanied by a slight performance drop.