To speed up the CREATE INDEX operation I increased the work_mem and set max_parallel_maintenance_workers to 8 and it worked as expected when I was creating a B-tree index on the table. I saw 7 parallel workers apart from the main one. But when I started to create a Hash index on the same table I saw that there was only 1 cpu core used by the only worker.

Are there some special options that I should change to gain the same result as when I was creating a b-tree index?

1 Answer 1


No. No one has implemented parallel build for hash indexes.

See the docs:

For index methods that support building indexes in parallel (currently, only B-tree)

So you would need to implement it before you could use it.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.