Agreement Protocols for a Dataspace System

Dataspace is the co-existence of a number of heterogeneous data sources which are not controlled by a single data source. Each data source has complete control over its data. The data in a dataspace come from various data sources distributed over various places, and these data may update periodically. Therefore, whenever a participant of dataspace get update, the changes must reflected in database if updated data is already populated. To provide stronger update guarantee in the dataspace is a challenging issue, and can be achieved by making an agreement among the owners of various data sources. The proposed work has presented the concept of agreement protocol suitable for a dataspace system. Our protocol defines two threshold values of data populated in the dataspace. On the basis of threshold values, the dataspace coordinator would decide whether the changes occur in data sources may get reflected on dataspace or not. The protocol is divided into a pull-based and push-based approach, also called reactive and proactive approach respectively.