Title: Data Synchronisation Techniques Model-driven Approach for Design and Execution of Applications on Multiple Clouds Members of the Modaclouds Consortium: Published Modaclouds Documents Modaclouds Model-driven Approach for Design and Execution of Applications on Multiple Clouds

NoSQL databases have emerged as the solution to handle large quantities of user-generated contents still guaranteeing fault tolerance, availability and scalability. Each NoSQL database offers differentiated properties and characteristics as well as different data models and architectures. As a result, the development of applications exploiting such kind of technology is strictly dependent on the specific NoSQL solution being adopted, and the migration from a NoSQL to the other requires the development of ad-hoc code managing the transfer of data. In order to mitigate such issue, this paper proposes an interoperable migration system for columnar NoSQL databases. The proposed approach is based on an orginal Metamodel, capable of preserving both strong and weak consistency between data updates, secondary indexes and various data types. Moreover, the approach allows developers to easily add support for new databases. Keywords-NoSQL databases; Data migration; Database as a Service (DaaS); Google App Engine Datastore; Azure Tables; BigTable; Amazon Dynamo; Amazon Dynamo DB