Re: Overhead for stats_command_string et al, take 2

Stefan Kaltenbrunner <stefan(at)kaltenbrunner(dot)cc> writes:
> This is what I get on a fast AMD Dual Opteron box(Running Debian
> Sarge/AMD64):

> 8.1.4 HEAD
> 1000000 SELECT 1; 74,74,73 77,76,77
> stats_command_string=1; 105,99,106 78,79,78
> log_min_duration_statement=100 79,80,81 75,80,76
> statement_timeout=100 78,79,78 75,79,77
> all 3 104,108,107 82,81,81

> all values in seconds with 3 consecutive runs of one million "SELECT 1;"
> queries. It takes about 48 seconds to run the same test without
> stat-collection btw.

I'm confused. Isn't your first table row for the case of no stat collection?
Or do you mean that you have stats_row_level and/or stats_block_level on
in all four cases?

regards, tom lane