If you are using OWL-DL reasoning (likely given the protege screenshot), there is a separate timeout you can set in your stardog.properties file: reasoning.schema.timeout
I was using SL reasoning option, so I changed to DL reasoning. However, the reasoning.schema.timeout = 120s config in the stardog.properties is not taken into account. The query using Stardog Studio never finish,I had to restart the server because the kill query option is not working.
But, the real problem is that the query with reasoning takes too long (more than 50 minutes before I have to restart), is there any way to improve reasoning? something like parallel reasoning, changing the reasoning type per query or clusterizing the database?
Reasoning with SL type, I get this error: QueryEval: com.complexible.stardog.plan.eval.ExecutionException: java.lang.RuntimeException: Running time of completion exceeded timeout of 60000
I'm attaching a zip file with the execution plan for RDFS and DL types. the query using RDFS returns the correct amount of records in 745 ms. The query using DL takes too long (after 50 minutes, I restarted the database server).
The best option for me will be to use SL reasoning, because I have some SWRL rules that use object properties with characteristics (symmetric, transitive, etc...). With RDFS the rules are correctly applied but I have no symmetric or transitive reasoning; with DL the query takes too long and no rules are applied; and with SL I get the error above.
Hi Stephen, anything detected in the query plans? It is possible that the java.lang.RuntimeException: Running time of completion exceeded timeout 60000 error could be solved?