Re: [PERFORMANCE] work_mem vs temp files issue - Mailing list pgsql-performance

From Tom Lane
Subject Re: [PERFORMANCE] work_mem vs temp files issue
Date
Msg-id 19928.1263245782@sss.pgh.pa.us
Whole thread Raw
In response to Re: [PERFORMANCE] work_mem vs temp files issue  (Jaime Casanova <jcasanov@systemguards.com.ec>)
List pgsql-performance
Jaime Casanova <jcasanov@systemguards.com.ec> writes:
> the temp files shoul be coming from hash operations but AFAICS the
> files are small and every hash operation should be using until
> work_mem memory, right?

No, when a hash spills to disk the code has to guess the partition sizes
(number of buckets per partition) in advance.  So it wouldn't be at all
surprising if the actual sizes come out substantially different from
work_mem.

            regards, tom lane

pgsql-performance by date:

Previous
From: Bob Dusek
Date:
Subject: Re: performance config help
Next
From: "Kevin Grittner"
Date:
Subject: Re: performance config help