Hi there,
my data model looks like this:
- V class Abstract with uuid (notuniqe index), model (lucene index) and
some other properties
- V class NormalizedNamedEntity with the same props and indexes like
Abstract
- E class uima_annotated which connects Abstracts with
NormalizedNamedEntities, and has also some other properties
Currently there are 100k Abstracs and 22k NormalizedNamedEntities and
nearly 2 million uima_annotated edges.
Now I'd like to perform queries like this:
select expand(out) from uima_annotated where in.uuid = "DBA002026" and
in.uuid = "NO000357"
This is like: give me all Abstracts (or the first 20) which have annotated
a DBA002026-NormalizedNamedEntity *AND* NO000357-NormalizedNamedEntity.
The query is not efficient and returns nothing, because "fetched more than
50000 records: to speed up the execution, create an index or change the
query to use an existent index".
I've tried to index uima_annotated.in, but this has no impact! Any ideas
how to speed this up? Where should I place an index? Is it possible to
index uima_annotated.in.uuid? Or is there a better way to express the query
(maybe coming from NormalizedNamedEntity)?
Thanks for any advice!
--
---
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 orient-database+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
