HYDRA is a SPARQL engine for querying SADI Semantic Web services, based on a scalable proprietary architecture, algorithms and data structures. It also implements simple federated querying of multiple SPARQL endpoints.
HYDRA can be used for data federation and/or semantic querying of data.
HYDRA can be used to query multiple distributed sources of data as a single database. Moreover, the data sources may be completely heterogeneous and include regular relational DB, online DBs, nomenclatures, ontologies, unstructured data such as scientific publication, Web services and various analytical programs. Practically any data that can be processed programmatically may be integrated with other resources. This (unprecedented) power of integration is achieved by wrapping the data sources as SADI Semantic Web services and orchestrating them in the engine to answer SPARQL queries.
HYDRA may be useful even in scenarios when only a few or even one database has to be queried because it supports semantic querying and is accessible to non-technical users. For example, if one has an SQL database, most non-technical users won't be able to run ad hoc queries over it without help from programmers or DBAs, because this requires knowledge of SQL and details of the specific data schema. Semantic queries, in contrast, are formulated in terms of the subject domain and thus are much more accessible to non-technical users, especially if they are supported with adequate user interfaces. Practically speaking, a database host can wrap his database as a number of SADI services, thus enabling its non-technical users to run multiple self-service ad hoc queries through HYDRA.
Published HYDRA case studies
- Interpreting ecotoxicology experiment data by discovering related information in multiple public databases.
Prior SADI case studies conducted with HYDRA's predecessor - SHARE - include:
- Integrating programs for structural analysis of small molecules with programs for ontology-based classification of molecules.
We have recently finished an alpha version of HYDRA (Aug 2013). Demos to interested parties can be arranged. The beta is expected in Jun-Jul 2014. Check this page if you would like to sign up to be a beta tester.
Check here what we can currently show.