-- Brian Baquiran <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I'm using Postgres 6.5.2 in a high-volume
> application that has on the order of a
> hundred inserts and selects (but mostly inserts) a
> minute. My loadaverage hovers
> around 1.0, and can go much, much higher (50-80)
> during vacuum and queries on
> fairly large tables (1M rows). This is on a 4-CPU
> Intel Xeon Linux machine with
> 1GB of memory. Is this load typical? How can I tune
> Postgres and my application
> to better handle the load?
>
> I'm running postmaster with the options -B600 -N300.
> I'm using the RedHat RPMs
> from a pretty-much-stock RH6.1 installation. Can I
> tune postgres to load more
> data into memory?
>
> Brian
Some suggestions:
1) Increase the -B and -S options
(I use -B5800 -S4000 )
2) Increase the default linux shared memory
(this may be necesary for the previous
item to work)
(for example, in the postgresql.init script
add the line:
echo "100000000" > /proc/sys/kernel/shmmax
)
3) Turn off flush to disk
(option -F )
(very little risk)
4) Use a 4kb disk-block size
instead of the default 1kb
(i'm not sure wheter this matters...)
( to look at the present size, try
dumpe2fs /dev/sda1 | grep size
where /dev/sda1 is the device where
postgresql lives)
5) Make a "vacuum analyze" daily
6) And the standard one: be careful with
the indexes. Adding (and also deleting)
an index may dramatically change the
performance.
Hope this helps...
Hernan Gonzalez
Buenos Aires, Argentina
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com
