What constitutes "reproducible" numbers from pgbench?

Поиск
Список
Период
Сортировка
От
Тема What constitutes "reproducible" numbers from pgbench?
Дата
Msg-id C5DBACC6DCC7604C9E4875FD9C7968B112D90C5000@ITXS01EVS.service.it.nrw.de
обсуждение исходный текст
Ответы Re: What constitutes "reproducible" numbers from pgbench?  (Qingqing Zhou <zhouqq.postgres@gmail.com>)
Re: What constitutes "reproducible" numbers from pgbench?  (Andy Colson <andy@squeakycode.net>)
Список pgsql-general
Hello list,
 
Exactly what constitutes „reproducible“ values from pgbench?  I keep getting a range between 340 tps and 440 tps or something like that using the same command line on the same machine.  Is that reproducible enough?
 
The docs state that one should verify that the numbers are reproducible, so I repeat any test run ten times before believing the results.  I’ve tried increasing the test duration (-T) from one minute to five minutes, then turning off autovacuum (in postgresql.conf) as recommended by the docs, but the range of results is not getting any narrower.  So what does “reproducible” mean as applied to pgbench?
 
Obviously I could be doing something wrong, such as missing some vital configuration option…
 
Thanks in advance for any insights.
 
Cheers,
Holger Friedrich
 

В списке pgsql-general по дате отправления:

Предыдущее
От: Craig Ringer
Дата:
Сообщение: Re: monitoring bdr nodes
Следующее
От: Qingqing Zhou
Дата:
Сообщение: Re: What constitutes "reproducible" numbers from pgbench?