Virtual Graph Materialization

Here are some observations on querying and materializing the triples received from a Postgres DB (these all use the same graph and the same WHERE clause in the body):

  • Query (select) output was pretty quick (I didn't note the time but it was pretty fast)
  • CONSTRUCT output was much slower but still ok (3 minutes)
  • INSERT timed out with a 20min timeout!

Just want to let you know ...

Andrea

Thanks Andrea. Were any of these run with reasoning? How many triples were returned from these queries? How many triples were in the database before the INSERT?

Jess, In reply ...

No reasoning.
CONSTRUCT returned about 100K triples, as I recall.
The base ontology was in the DB (13K triples) before the INSERT.

Andrea

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.