Data heterogeneity is a major issue in any context where software directly deals with data. The most general expectation of any complex system is the so-called seamless integration, where data can be accessed, retrieved and handled with uniform techniques, tools and algorithms.

The aim of this work is dealing with data heterogeneity and data integration techniques under a number of perspectives.

From the theoretical perspective the core problem of heterogeneity is that data can be intrinsically different because multiple data models are adopted to organize them. Here model management is considered as the framework to formalize model and data translation problems: a schema, instance of a certain model is translated to another schema instance of a target model relying on a model-independent approach based on a general meta-level.

From the performance perspective translations cannot be performed out of the involved systems with an import-translate-export process. The schema and data translation approach has been extended in order to perform runtime translations and automatically generate views of data.

As application example, a model-independent solution to the round-trip engineering problem is illustrated, showing the typical propagation of changes among related schemas.

Nowadays market demand for highly specialized data processors, performing at best in specific cases such as web content retrieval, document search, object serialization, parallel calculation is taken in particular consideration. NoSQL engines promise exceptional performance in non transactional fields and leverage simplified but peculiar data models. Therefore a core goal of data integration is providing techniques and tools to facilitate the interaction with these systems from both a theoretical and technical perspective. A new interface has been defined having as goal to support applications by hiding the heterogeneity of the languages and the interfaces of the various NoSQL systems.