Re: Slow query, where am I going wrong?

Albe Laurenz *EXTERN* wrote
> Increasing the statistics for test_result.id_recipe_version
> had no effect?
>
>> I am going to normalize the table some more before partitioning.
>
> How do you think that partitioning will help?

I increased the statistics in steps up to 5000 (with vacuum analyse) - Seems
to be as good as it gets.

http://explain.depesz.com/s/z2a

The simulated data is about a months worth. Partitioning is only really
expected to help on insert, but that's pretty critical for us.

At the moment test_item contains way too much repeated data IMHO, and I will
address that asap (is going to hurt ).

I will also look into creating an aggregate table to hold the 'distinct'
values.

Andy.





--
View this message in context:
http://postgresql.1045698.n5.nabble.com/Slow-query-where-am-I-going-wrong-tp5730015p5730140.html
Sent from the PostgreSQL - performance mailing list archive at Nabble.com.