Page MenuHomePhabricator

[Story] Decide the back-end storage engine
Closed, ResolvedPublic

Description

Propose to the community a set of solutions to ensure the self-sustainability of the back-end storage engine.
The current implementation is in SQL.

Possible alternatives to store Wikidata items, ordered by priority:

  1. Blazegraph, as a triple store accepting the RDF serialization;
  2. MongoDB, as a document store accepting the JSON serialization.

Event Timeline

Hjfocs renamed this task from [Story] Decide the dataset storage to [Story] Decide the back-end storage engine.May 29 2017, 11:25 AM

Blazegraph with datasets serialized in RDF and following the Wikidata RDF data model seem to be the most standard way: RDF is a global Web standard and also enables query capabilities that will be needed to implement both the domain and the SPARQL filter, cf. T166512.