Re: High CPU load on Postgres Server during Peak times!!!!

Поиск
Список
Период
Сортировка
От Merlin Moncure
Тема Re: High CPU load on Postgres Server during Peak times!!!!
Дата
Msg-id b42b73150910090627i30762e26ve695ae8305982bc5@mail.gmail.com
обсуждение исходный текст
Ответ на Re: High CPU load on Postgres Server during Peak times!!!!  (Shiva Raman <raman.shivag@gmail.com>)
Список pgsql-performance
On Fri, Oct 9, 2009 at 3:11 AM, Shiva Raman <raman.shivag@gmail.com> wrote:
> Dear all
>   with reference to the discussions and valuable suggestions i got from the
> list, the code has been reviewed and updated with explicit commit . There is
> a good improvement in  performance .I am also planning to upgrade the
> database from 8.1 to 8.3 /8.4 .
>    My current OS is SLES 10 SP3 default ships with postgresql 8.1 .
>   The latest version of SLES 11 ships with postgresql 8.3 version.
> I will be upgrading the Postgersql on my SLES 10 SP3 for PPC only.
> I have not seen any prebuilt RPMS of Postgresql 8.3 or 8.4 version for SLES
> 10  PPC architecture .
> When I tried to build the PPC RPM from Source in the PowerPC, it shows lot
> of dependancies.
>    I have decided to install from source - Postgres 8.3 / Postgresql-8.4.
> tar.gz .
>
> Is there any major changes/updates in my 8.1 database  i have to take care
> while  upgrading to postgresql 8.3/ 8.4 ?  Is 8.3 or 8.4 the right version
> to upgrade from 8.1 ?
>
> Please let me know your suggestions.


The 'big picture' issues:
*) Test your postgresql.conf first. Some settings have changed or have
been removed (like fsm).
*) Many implicit casts to text were removed. Essentially the server is
less tolerant of sql that many would consider buggy
*) autovacuum is now on by default

and, most importantly:
*) sit back and enjoy the speed :-)

regarding 8.3/8.4, it's a tough call.  8.4 has a better chance of
being supported by in place upgrade in the future, so i'd start there.

merlin

В списке pgsql-performance по дате отправления:

Предыдущее
От: Kenneth Marshall
Дата:
Сообщение: Re: Bad performance of SELECT ... where id IN (...)
Следующее
От: Alan McKay
Дата:
Сообщение: disk I/O problems and Solutions