rmagen
rmagen
Theoretically we should be able to run distributed Unipop queries on Spark or something. Some of Unipop's data sources even have Hadoop integration (e.g. Elasticsearch RDD, Jdbc RDD, etc). Utilizing...
Most database optimizers use statistics-based cardinality estimates to to determine the optimal order in which to run a query's steps. Should we do something similar? We could implement a `TraversalStrategy`...
Expected results: [{a=v[340], b=v[258], c=v[228], d=v[592], e=v[749]}] Actual results: []
The current "siblings" implementation is kind of cumbersome, and not very optimized in certain use-cases. Try to do the bulking in the steps. https://issues.apache.org/jira/browse/TINKERPOP3-702
Apparently the ES Scroll API is not supposed to be used for OLTP situations. https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-scroll.html Should we keep it as-is, or revert to regular queries, or enable both options?
When searching for vertices/edges, we should only fetch the id, without the document source with all the fields. the source should be fetched only when asked for. We should probably...
the DefaultSchemaProvider currently creates a sub-optimized index. - make the default analyzer "not_analyzed", instead of "keyword". http://www.elastic.co/guide/en/elasticsearch/reference/current/mapping-root-object-type.html - disable the "_all" field.