How to optimize the performance of scanning edge data

[czp]

I have a graph that has 80 million vertexes and 30 million edges. I need to scan all the edge data, including the ids of its starting and ending vertexes.But the performance is poor and I need to optimize it.

This is my query:

MATCH (v)-[e:E1]->(n) RETURN id(v) AS starting, id(n) AS ending, e.prop1, e.prop2,...

Can anyone give some ideas on optimization or possible solutions for this scenario?

[czp]

Neo4j Version: 3.5

Operating System: Ubuntu 20.04 / Docker

API: Cypher

[glilienfield]

The query is fairly straight forward. Are you trying to return millions of rows of data in neo4j browser? If so, maybe try cypher-shell instead.

https://neo4j.com/docs/operations-manual/current/tools/cypher-shell/

[czp]

I'm using a cypher-shell. Is there any other way to optimize it?

[czp]

I tried relationship index, but version 3.5 doesn't support it.

[czp]

https://neo4j.com/developer-blog/neo4j-4-3-blog-series-relationship-indexes/

[glilienfield]

I don’t think relationship indexes would help since you are not searching on a relationship property. How about you determine the query plan and post it, v