cancel
Showing results for 
Search instead for 
Did you mean: 

How can I optimize my Cypher Queries?

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?

1 ACCEPTED SOLUTION

There are a number of things you can do:

  1. Run your query with PROFILE or EXPLAIN. In the resulting query plan there are indications of expensive operations.
  2. Make sure that lookup operations by id or value (numeric range, text search) run on fields that are indexed.
  3. If you are running updates make sure that your updates handle batches of about 10k-100k records, if you have more, please batch them.

Share your queries and query plans (download and share the PNG) in this #neo4j-graph-platform:cypher category. Then others can help you.
It is also helpful to share your data-model and other statistics, and also the actual Cypher query.

There are already a number of performance related knowledge base posts available.

You can learn more here:

Cheers, Michael

View solution in original post

4 REPLIES 4

There are a number of things you can do:

  1. Run your query with PROFILE or EXPLAIN. In the resulting query plan there are indications of expensive operations.
  2. Make sure that lookup operations by id or value (numeric range, text search) run on fields that are indexed.
  3. If you are running updates make sure that your updates handle batches of about 10k-100k records, if you have more, please batch them.

Share your queries and query plans (download and share the PNG) in this #neo4j-graph-platform:cypher category. Then others can help you.
It is also helpful to share your data-model and other statistics, and also the actual Cypher query.

There are already a number of performance related knowledge base posts available.

You can learn more here:

Cheers, Michael

clem
Graph Steward

Here's an interesting trick to optimize a query. It involves manually breaking up the query into two parts and forcing the part that winnows down the results be executed first returning a small result, and then executing a potentially slow query on the resulting small set.

clem
Graph Steward

Here's a useful item on Relationships or Relationship Properties;

http://www.dominickumar.com/blog/neo4j-relationship-modelling-performance/

Awesome, thats how traversing with index free adjacency.
Just a small thought, even with having property on relationship along with manual index on relationship property (by using **CALL db.index.fulltext.createRelationshipIndex)
could you co-relate the performance metrics