With the rapid development of hand-held mobile devices and Big Data, the issues of data storage have attracted more and more attention. Dynamic replication strategy with high reliability, low cost and automatic fault tolerance has become the common goal of both sides of the service. However, this kind of data replication not only consumes a large amount of storage resources, but also the problems of system failure, access latency and load grow with the expansion of the system. In this paper, a dynamic distribution strategy of replicas based on multi-objectives optimization is proposed to optimize the three objectives such as file availability, response time and load balancing. The unique artificial immune algorithm is used to determine the optimal distribution of replicas. The multi-objectives optimization is considered when our strategy dynamically assigns the replicas to reasonable storage nodes. Simulation results show that the proposed strategy can seek the near optimal solutions by balancing the trade-offs and effectively improving reliability, reducing delay and optimizing load balance.