Authors:
Eva Kühn
;
Angelika Ruhdorfer
and
Vesna Šešum-Cavic
Affiliation:
Vienna University of Technology, Institute of Computer Languages, Austria
Keyword(s):
Asynchronous replication, conflict detection and resolution, relational database, coordination space, data grid.
Related
Ontology
Subjects/Areas/Topics:
Business Analytics
;
Communication and Software Technologies and Architectures
;
Data Engineering
;
Data Warehouses and Data Mining
;
e-Business
;
Engineering Information System
;
Enterprise Information Systems
;
Information Systems Analysis and Specification
Abstract:
Data replication is a well-known technique in distributed systems, which offers many advantages such as higher data availability, load balancing, fault-tolerance, etc. It can serve to implement data grids where large amounts of data are shared. Besides all advantages, it is necessary to point to the problems, called replication conflicts that arise due to the replication strategies. In this paper, we present an infrastructure how to cope with replication of heterogeneous data in general for conflict detection and resolution and we illustrate its usefulness by means of an industrial business case implementation for the domain of relational databases and show further extensions for more complex resolution strategies. The implementation deals with the special case of asynchronous database replication in a peer-to-peer (multi-master) scenario, the possible conflicts in this particular domain and their classification, the ways of conflict detection, and shows some possible solution methods.