A dynamic data allocation method with improved load-balancing for cloud storage system

With the growing demand of data and the increase of the user scale, data allocation technology has become a key technology for improving scalability and flexibility in current mass storage system such as cloud storage system. This paper proposed an efficient dynamic data allocation strategy with data partitioning and load balancing. Based on the basic idea of consistent hashing algorithm, the strategy introduced the concept of virtualization technology and improved the load-balance with employing virtual node. Moreover, the strategy adopted a novel available-storage-capacity-aware and storage-capacity-utilization-aware method to enhance the performance of the cloud storage system. The simulation results demonstrate that the proposed data allocation strategy improves system performance in both homogeneous and heterogeneous distributed storage architectures.