AutoMed: Data Integration using Both as View (BAV) Rules
AutoMed is a framework and software package comprising of graphical
tools, and a programmers API, that provides new solution to the
problem of model based data integration. Particular features of
AutoMed that distinguish it from other available data integration
products are:
- AutoMed is based on a data integration approach called both as
view (BAV). BAV provides a richer integration framework
than is provided by GAV, LAV, and GLAV approaches. In particular, it
is capable of providing a complete mapping between schemas in both
directions, and hence is of particular relevance to P2P applications
and Web Services.
- AutoMed is capable of handling a wide range of data sources, with
the current implementation supporting data held relational DBMSs, XML
files, and structured flat files, and development work on unstructured
text files underway. By basing the representation of these various
models on an underlying nested hypergraph model called the HDM,
AutoMed is also capable of automating the translation between these
data models. A further advantage of basing the approach in the HDM is
that tools such as the graphical editor and the schema matching tool
are capable of operating over any structured or semi-structured data
model after a very simple configuration of the AutoMed toolkit.
- AutoMed separates the logical and operational aspects of data
integration. A BAV pathway specifies the logical relationship between
schemas. It is left as an operational matter to choose which schemas
relate to data sources, and which are virtual. This flexible approach
allows AutoMed to be used as the basis for distributed databases
(based on virtual global schemas derived from several data sources),
data warehouse environments (based on materialised global schemas derived from
several data sources), and schema evolution (where a materialised schema
may change to a virtual schema, derived from the new version of the data source
as a new materialised schema).
AutoMed was originally an EPSRC funded project, but development work continues
at Birkbeck and Imperial Colleges, under a number of related projects.
AutoMed Webmaster: Peter McBrien (p.mcbrien@imperial.ac.uk)