Samantha Bail, Sandra Alkiviadous, Bijan Parsia, David Workman, Mark Van Harmelen, Rafael S. Goncalves, and Cristina Garilao: FishMark: A linked data application benchmark. In: Proc. of SSWS+HPCSW 2012.
We provide the data and queries used for the benchmark, as well as the MUM-benchmark framework.
- The benchmarking framework is based on the Berlin SPARQL Benchmark framework).
- The mappings for the D2R translation from relational data into RDF are provided by the FishDelish project.
- The FishBase data has been provided by the FishBase Consortium.
- can be found on: http://code.google.com/p/mum-benchmark/
- can be used with the FishMark data (relational data and RDF) and the FishMark queries
- can also be used with new data and queries to create an application benchmark
- contains query templates (in data/example) for the FishMark application benchmark
- is Open Source, and contributions/extensions are welcome.
- can be downloaded from: http://code.google.com/p/mum-benchmark/downloads/list
- The query set consists of the FishMark query templates in SQL and SPARQL format,
- an OWL ontology
- and an OBDA mappings file
The FishBase data contains:
- a complete dump of the fishbase.org MySQL database
- mapping files for the D2R tool to generate
- a complete translation of the SQL data into RDF (~1.38bn triples)
- a minimal translation of the SQL data into RDF (~300m triples, approx. 30,000 species) which is sufficient for the queries
- 3 RDF datasets based on the minimal translation, scaled by the number of species
The FishBase data available for download on this site is provided for testing purposes only. As the data is provided without the context of the fishbase.org application, we cannot guarantee that any information extracted from it is correct outside the given context. Therefore, the data must not be used in a production environment or any application which provides information about fish. Please be aware of this when downloading the data.