For reference, our test graph has 300,000 vertices and 1.5 million edges
and we are using 2.0-SNAPSHOT. See the example schema and queries below:
*(schema)*- Attraction, PostalCode, City, and Concepts are vertices
- isLocatedInPostalCode, belongsInCity, and hasConcept are Edges
- cypher:
(ATTRACTION)-[:isLocatedInPostalCode]->(POSTALCODE)-[:belongsInCity]->(CITY)
- cypher: (ATTRACTION)-[:hasConcept]->(CONCEPT)
*(query)*SELECT
FROM Attraction
WHERE out("isLocatedInPostalCode").out("belongsInCity").name IN ["Seattle",
"Austin"] AND out("hasConcept").name IN ["spa", "pool"]
This query works, but is very slow. It is also not using our indexes that
are on City.name or Concept.name
*Our goal here is to find all Attractions from a specific cities that also
has specific concepts. * This requires traversing out from Attraction down
the location path (Attraction to PostalCode to City) *AND* down the concept
auth (Attraction to Concept). So far, we seem to be unable to do this type
of query that makes uses of indexes and is fast. Note, we have other
similar traversals we need to do from Attraction in this one query (other
paths)... this is just an example with two.
In cypher / neo4j, this type of thing is easy (and uses indexes), just
wondering how to do this with OSQL.
Thank you in advance.
Brandon
--
---
You received this message because you are subscribed to the Google Groups
"OrientDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
