Abstract: File replication is a common strategy to improve data reliability and availability in large clusters. Reliability
for each file under server failures based on the relationship between file reliability and replication factor when servers
have a certain probability to fail. In the existing system more number of replica creates it require more energy
consumption, time and storage space and sometime system fails and cannot send immediate response to the user request.
In the proposed system increasing number of file replica according to user request or user priority and vice versa.
Propose strategies in reducing file read latency, replication time and power consumption in large cluster. If multiple user
send request for one file then system create multiple replica according to priority and get immediate response to the user.
Keywords: Data-intensive clusters, file replication, replica placement, energy-efficient