A Dynamic Replica Placement Mechanism Based on Response Time Measure

Most of existing replication strategies mainly distribute replicas based on access frequency or total response time of the whole system, lacking of consideration of the response time requirement of a single request. To overcome the deficiencies, a model called RPRTM (Replica Placement based on Response Time Measure) is presented, and a genetic algorithm is designed to figure out the model. Example analysis indicates that the RPRTM can satisfy every node’s response time requirement of a single request and minimize the number of replicas at the same time. In addition, it reduces the total response time at most to improve the whole system performance.