2 Replies Latest reply on Feb 15, 2016 2:57 PM by Matperry-Oracle

    Property path query performance problems: help desperately needed.

    07291868-dafd-4ac1-a608-b4a39a6efb1f
    07291868-dafd-4ac1-a608-b4a39a6efb1f Feb 12, 2016 2:52 AM

      I am new to OracleRDF but not new to triple stores or SPARQL

       

      I have some existing queries that run on various other triple stores (Virtuoso, Allegrograph, Sesame, Jena memory model etc). The graph I am testing against only contains 1600 triples (yes sixteen hundred)

       

      However even simplified queries with property paths either never return or throw a java heap problem (I have 4Gbyte allocated to the Tomcat that is running Joseki)

       

      PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#>

      PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>

      PREFIX owl: <http://www.w3.org/2002/07/owl#>

      PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>

      SELECT  distinct *

                     WHERE {

                      {

                                      ?property rdfs:subPropertyOf* ?sproperty .

                                     ?sproperty rdfs:domain ?sdomain .

                                      ?sproperty rdf:type ?propertyType .

                                      ?property rdfs:range ?r .

                     }

      }

       

      i have tried every combination of hints to get this query to work. For example:


      PREFIX ORACLE_SEM_FS_NS: <http://oracle.com/semtech#TIMEOUT=100,ASSERTED_ONLY,JENA_EXECUTOR>


      The problem is this is not even close to the complexity of the queries i really need to run (and already run on 4 different triple stores without any issue).


      I have looked at all of the hints available and tried most permutations.


      What do I do other than switch to a different triple store?


      Help!!!