Incorrect results for .count() when using mixed index and trying to count a large amount of nodes
Using a Lucene mixed index when counting nodes has inconsistent behavior depending on the amount of nodes counted.
When I expect a result of >100k, Janusgraph returns a value somewhere between 1000-4000 instead of the correct result.
The only way I've been able to force it to always fully count the nodes is by replacing .count() with .constant(1).count() in all my queries, which works but seems inefficient. Does anyone have any tips on how to avoid doing this / improve performance of large .count() queries?
Thank you!
Fyi: JanusGraph version is 1.1.0 with mostly default settings. query.smart-limit is unspecified, which should default to false.
0 Replies