How and where are the underlying data stored in PoolParty?

For every PoolParty project and for every reference corpus, a Sesame repository will be generated. Each project repository consists of various named graphs:PoolParty data is stored in a Sesame triple store. Usually the project data is kept in a 'Memory Repository', which is lightning fast. Please note that if you have many projects using 'Memory Repositories', then your PoolParty server needs sufficient memory (in most cases 16GB RAM is enough). As an alternative, you can select 'Native Repository'. In this case, with larger projects (more than 5,000 concepts) the performance might suffer from rather poor performance. You can select the type of repository during project creation: see the documentation.

  • the default graph, which stores the actual thesaurus data

  • the metadata graph,

  • a workflow graph in case workflows have been activated for the project,

  • a history graph, which stores all events per concept, and

  • one graph per linked data source to which mappings have been made and data has been harvested from (in order to keep track of the provenance)