Re: The same query - much different runtimes

Поиск
Список
Период
Сортировка
От Pavan Deolasee
Тема Re: The same query - much different runtimes
Дата
Msg-id CABOikdPQqi5+DgV_CGuxG91cMojQX7ZQ9sxNx7SXKSiER_EXbg@mail.gmail.com
обсуждение исходный текст
Ответ на The same query - much different runtimes  (Johann Spies <johann.spies@gmail.com>)
Список pgsql-performance
Дерево обсуждения

On Mon, Apr 7, 2014 at 3:55 PM, Johann Spies <johann.spies@gmail.com> wrote:


I then ran the query and the result was produced in about the same time as (2945 ms).

What can cause such a huge discrepancy?  

May be when you reran the query, most of the data blocks were cached either in the shared buffers or the OS cache. That could drastically improve the performance. I can see a large number of shared buffer hits in the explain analyze output of the query ran through psql session.

Thanks,
Pavan

--
Pavan Deolasee
http://www.linkedin.com/in/pavandeolasee

В списке pgsql-performance по дате отправления:

Предыдущее
От: Ryan Johnson
Дата:
Сообщение: Re: SSI slows down over time
Следующее
От: Cora Ma
Дата:
Сообщение: performance degradation after launching postgres cluster using pgpool-II