New Ph D student exploring graph query optimization

[amelia]

Hello! I'm Amelia, a new Ph. D. student at Colorado School of Mines in the USA, and plan to do my research in graph database query optimization ... I'm starting out by exploring the capabilities of neo4J from a user perspective, and then plan to dive into the opensource libraries. I have a background in SQL database systems internals (DEC/RDB, HP/Allbase, Sybase SQL Server, Cloudscape/Apache Derby) and am looking forward to exploring this new-to-me space!

[clem]

There's a few of threads on optimization:

Slow query with very limited data and boolean = false Cypher

Happy to report I may have solved the issue! To limit the siblings that get searched for reserved = false, I set up a subquery in the CALL block. CALL { MATCH (origin { type: 'part', identifier: '182' })<-[:CHILD]-(parent)-[:CHILD]->(siblings) RETURN siblings } WITH siblings WHERE siblings.reserved = false RETURN siblings The query planner shows this searching about ~475 nodes for the boolean attribute, instead of ~50,000.

How can I optimize my Cypher Queries? Cypher

What should I do if my Cypher query is slower than expected, or uses more memory or CPU than I have? I'm running my queries with Neo4j Browser, copied from my application. What should I do? Are there any good practices / pointers?

Definitely learn about EXPLAIN and PROFILE (which I suspect you already are familiar with since they are in other DBs.)

Since the code is open source, you could always see how Neo4J implements its query optimizations.

Good luck with your PhD!

[dominicvivek06]

Hello @amelia , since you are coming from SQL Database system internals, you will find similar Query profiling in neo4j also (with neo4j terminilogy) , and the keywords -> predicates, filters, merge and cartesian joins do exists in neo4j also.