Currently, most joins in Comunica will go via the Bind Join algorithm, as this can handle federation over heterogeneous source quite nicely through adaptive query planning.
In certain cases, existing query planning algorithm may however provide better performance (e.g. for #548), so we should try implementing different query planning algorithms as join actors, and evaluate their performance.
Possible planning algorithms:
Haas, Laura, et al. "Optimizing queries across diverse data sources." (1997).
Acosta, Maribel, and Maria-Esther Vidal. "Networks of linked data eddies: An adaptive web query processing engine for RDF data." International Semantic Web Conference. Springer, Cham, 2015.
Galkin, Mikhail, et al. "Smjoin: A multi-way join operator for sparql queries." Proceedings of the 13th International Conference on Semantic Systems. 2017.
Heling, Lars, and Maribel Acosta. "Robust Query Processing for Linked Data Fragments."
Issue type:
Description:
Currently, most joins in Comunica will go via the Bind Join algorithm, as this can handle federation over heterogeneous source quite nicely through adaptive query planning.
In certain cases, existing query planning algorithm may however provide better performance (e.g. for #548), so we should try implementing different query planning algorithms as join actors, and evaluate their performance.
Possible planning algorithms: