Skip to Main Content
Data-intensive application within cloud environment needs huge data transfer and migration. In order to optimize the use of data resource, improve the reliability of data and reduce data-access latency and bandwidth consumption, a replica pre-adjustment strategy based on trend analysis of file popularity is proposed. This replica strategy adaptively and dynamically adjust the number of replicas of a file by analyzing the popularity information of the file and comprehensive considering the load balancing and utilization ratio of storage space, which makes the number and distribution of replicas more reasonable, and minimize the costs and maximize the profit of a replica. Experimental results indicate that the algorithm dramatically shortened the execution time of a job and avoided excessive consumption of storage space and decrease overall bandwidth utilization.