Architecture of the multiagent system for replicated data management

Abstract

Data replication is one of the ways to fulfil growing expectations of computer systems end users. It allows to improve reliability and performance of software solutions. On the other hand, multiagent systems is a computation model which may facilitate an exploit of large distributed systems potential. Combination of both mechanisms gives new possibilities in such systems building. Architecture of one of these systems is presented in the paper. The main goal of the system is usage of various types of agents in data replication process. Its implementation was realised using the JADE environment. © 2011 Springer-Verlag Berlin Heidelberg.