Re: [HACKERS] database size

Поиск
Список
Период
Сортировка
От The Hermit Hacker
Тема Re: [HACKERS] database size
Дата
Msg-id Pine.NEB.3.96.980106203122.254f-100000@thelab.hub.org
обсуждение исходный текст
Ответ на Re: [HACKERS] database size  (darrenk@insightdist.com (Darren King))
Ответы Re: [HACKERS] database size  (Bruce Momjian <maillist@candle.pha.pa.us>)
Список pgsql-hackers
On Tue, 6 Jan 1998, Darren King wrote:

> 48 bytes + each row header (on my aix box..._your_ mileage may vary)
>  8 bytes + two int fields @ 4 bytes each
>  4 bytes + pointer on page to tuple
> -------- =
> 60 bytes per tuple
>
> 8192 / 60 give 136 tuples per page.
>
> 300000 / 136 ... round up ... need 2206 pages which gives us ...
>
> 2206 * 8192 = 18,071,532
>
> So 19 MB is about right.  And this is the best to be done, unless
> you can make do with int2s which would optimally shrink the table
> size to 16,834,560 bytes.  Any nulls in there might add a few bytes
> per offending row too, but other than that, this should be considered
> normal postgresql behavior.

    Bruce...this would be *great* to have in the FAQ!!  What we do need is
a section of the User Manual dealing with computing resources required for
a table, similar to this :)

Marc G. Fournier
Systems Administrator @ hub.org
primary: scrappy@hub.org           secondary: scrappy@{freebsd|postgresql}.org


В списке pgsql-hackers по дате отправления:

Предыдущее
От: The Hermit Hacker
Дата:
Сообщение: Re: [HACKERS] database size
Следующее
От: Keith Parks
Дата:
Сообщение: Another regression test failure.