Want: in hash

[clem]

I believe one thing that can cause degradation in performance, is that the IN LIST operator does a linear search across an array (maybe I'm wrong...).

Perhaps Neo4J could turn the list into a hash before execution so that testing for membership is faster.

The documentation doesn't say otherwise. If Cypher does treat a LIST as a hash when doing a membership test, it should say so in the documentation.

See:

Query optimization that collects and orders nodes on very large graph Cypher

I have a decently large graph (1.8 billion nodes and roughly the same number of relationships) where I am performing the follow query: MATCH (n:Article) WHERE n.id IN $pmids MATCH (n)-[:HAS_MENTION]->(m:Mention) WITH n, collect(m) as mentions RETURN n.id as pmid, mentions ORDER BY pmid where $pmids are a list of strings, e.g. ["1234", "4567"] where the length of this list varies from 100-500 length. I am currently am holding the data within neo4j docker community instance with the following c…

[michael_hunger]

For IN LISTit is actually turned into a (hash)-set on the first iteration so subsequent uses of that list use the companion hashset internally.

if there is an index on :Article(id) it will send the whole list to the index instead doing an IN LIST

So it should be reasonably fast if there is enough RAM and/or fast IO.

[clem]

This needs to be mentioned in the documentation!

Thanks.