A Framework for Data Modeling and Querying Dataspace Systems

A dataspace system manages the large scale heterogeneous collection of data distributed over various data sources in different formats. It addresses the structured, semistructured, and unstructured data in coordinated manner without presuming the semantic integration among them. Therefore, the management of heterogeneous data in a dataspace system is a crucial challenge. This work introduces a layered framework to investigate the data management issue in a dataspace system. Our framework is based on the dataspace principle, i.e., “Pay-As-You-Go principle”, which improves the performance of the system with time, and adds the efforts when require. We have discussed several research issues for designing of a dataspace system.