mirror/postgresql
2
0
Fork 0
mirror of https://git.postgresql.org/git/postgresql.git synced 2024-12-15 08:20:16 +08:00
Code Issues Packages Projects Releases Wiki Activity

pgbench: Correct for bias in --rate schedule generation.

Browse Source 
Previous code gave a mean delay 0.44% below target.  This change also
has the effect of increasing the maximum possible delay.

Fabien COELHO
This commit is contained in:
Noah Misch 2013-09-25 23:09:50 -04:00
parent 77ae7f7c35
commit c2df45a37c
1 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
contrib/pgbench/pgbench.c
View File

@ -929,13 +929,17 @@ top:
* that the series of delays will approximate a Poisson distribution * that the series of delays will approximate a Poisson distribution
* centered on the throttle_delay time. * centered on the throttle_delay time.
* *
* 1000 implies a 6.9 (-log(1/1000)) to 0.0 (log 1.0) delay multiplier. * 10000 implies a 9.2 (-log(1/10000)) to 0.0 (log 1) delay multiplier,
* and results in a 0.055 % target underestimation bias:
*
* SELECT 1.0/AVG(-LN(i/10000.0)) FROM generate_series(1,10000) AS i;
* = 1.000552717032611116335474
* *
* If transactions are too slow or a given wait is shorter than * If transactions are too slow or a given wait is shorter than
* a transaction, the next transaction will start right away. * a transaction, the next transaction will start right away.
*/ */
int64 wait = (int64) int64 wait = (int64) (throttle_delay *
throttle_delay * -log(getrand(thread, 1, 1000)/1000.0); 1.00055271703 * -log(getrand(thread, 1, 10000)/10000.0));
thread->throttle_trigger += wait; thread->throttle_trigger += wait;