Обсуждение: queries on huge tables

Поиск
Список
Период
Сортировка

queries on huge tables

От
"Lending, Rune"
Дата:
Hello all.
 
I am having a couple of tables with couple of hundre millions records in them. The tables contains a timestamp column.
I am almost always interested in getting datas from a specific day or month. Each day contains aprox. 400.000 entries.
 
When I do such queries as " select ... from archive where m_date between '2005-01-01' and '2005-02-01' group by ... " and so on.
It takes very long. I am having indexes that kicks in, but still it takes sometime.
 
I have splitted the archive table in smaller monthly tables, it then goes a lot faster, but not fast enough.
 
I know simular systems that uses Oracle and gains a lot on performance because of the partioning. That kind of anoyes me a bit :)
 
Does anyone of you have some good ideas on how speed up  such queries on huge tables?
 
regards
rune
 
 
 
 

Re: queries on huge tables

От
Mike Rylander
Дата:
The most recent version of this thread starts here:
http://archives.postgresql.org/pgsql-general/2005-03/msg00321.php .
Search the archives for "table partition", "union view" and "partition
inherits" and you should find most relevant discussions.

Hope that helps!

On Thu, 17 Mar 2005 15:01:43 +0100, Lending, Rune
<rune.lending@inpoc.com> wrote:
>
> Hello all.
>
> I am having a couple of tables with couple of hundre millions records in
> them. The tables contains a timestamp column.
> I am almost always interested in getting datas from a specific day or month.
> Each day contains aprox. 400.000 entries.
>
> When I do such queries as " select ... from archive where m_date between
> '2005-01-01' and '2005-02-01' group by ... " and so on.
> It takes very long. I am having indexes that kicks in, but still it takes
> sometime.
>
> I have splitted the archive table in smaller monthly tables, it then goes a
> lot faster, but not fast enough.
>
> I know simular systems that uses Oracle and gains a lot on performance
> because of the partioning. That kind of anoyes me a bit :)
>
> Does anyone of you have some good ideas on how speed up  such queries on
> huge tables?
>
> regards
> rune
>
>
>
>


--
Mike Rylander
mrylander@gmail.com
GPLS -- PINES Development
Database Developer
http://open-ils.org