An Improved Method for Consistent Replicated Data in the Intelligent Network


The Intelligent Network (IN) is a novel architectural concept for telecommunication networks that enables network operators as well as independent service providers to swiftly introduce new services into networks. Furthermore, these services can be made flexible enough so that after deployment, service subscribers can directly tailor them to their requirements. The main idea behind the IN concept is the separation of switching functionality from service control. In order to achieve a high degree of flexibility the service logic is realized by software modules called service logic programs which can be customized with subscriber specific data. Thus, a large amount of data (service logic programs, subscriber specific data and management information) needs to be handled. As telecommunication systems are of highly parallel nature the main problem for the database system is to provide consistent data for all these service logic programs running at the same time. Using a centralized DBMS to control data consistency will rapidly lead to a bottleneck. Therefore you have to deal with distributed databases having the problem of heavy communication to provide data consistency. If you intend to replicate data (fully or partially) over the whole distributed system due to performance reasons and reliability then the communication overhead further increases because of the synchronization of the replicas. We developed a concept for the specific needs of the IN. It can be seen as a replicated database system having lower communication overhead than classical distributed database systems. It also impresses by its ease of scalabilig. In this talk we will first look at the IN architecture, then discuss our concept in detail and finally show how it can be integrated into the IN. 0-7803-18 11-0/94$04.00

DOI: 10.1109/NOMS.1994.643309

Cite this paper

@inproceedings{Gallersdrfer1994AnIM, title={An Improved Method for Consistent Replicated Data in the Intelligent Network}, author={Rainer Gallersd{\"{o}rfer and Karin Klabunde and Ralf Nellessen}, booktitle={NOMS}, year={1994} }