Data replication as a tool to improve the reliability of distributed systems

The problem of improving the reliability of distributed data processing systems is considered. A method for replicating data sets is presented as an effective tool for ensuring a high level of data safety in distributed systems. Examples of methods of static and dynamic replication are given. The formulation and algorithm of the problem of finding the optimal distribution of replicas over nodes of a computer network is given. An example of operation of the proposed algorithm for solving the formulated problem is presented. The analysis of the results obtained by the algorithm are analyzed.