Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: after adding 6 new datafiles the performance go down

Re: after adding 6 new datafiles the performance go down

From: Ganesh Raja <ganesh_at_gtfs-gulf.com>
Date: Wed, 20 Feb 2002 12:47:02 +0400
Message-ID: <eho67ukemjrvom49g6rp6h0tqrphmj5laa@4ax.com>


I Feel there is ...

When u checkpoint the DBWR Fushes the Dirty Blocks to the buffer when the DBWR Flushes the Dirty Blocks the log buffer is also flushed into the Redo Logs. If the Log Buffer is very big then this will take more time... This is no way related to theproblem stmt but stil ....

On Sun, 17 Feb 2002 14:45:00 GMT, "koert54" <koert54_at_nospam.com> wrote:

>>>If you are doing excessive checkpoints you may try
>>>increasing the size of log_buffer.
>
>hmmm - this is a new one ! :-)
>I know log_checkpoint_interval, log_checkpoint_timeout, increase logfile
>members and the number of logfile groups...
>but as far as I know there's no relationship between the frequency of
>checkpointing and log_buffer !
>
>
>"Stephen Ashmore" <sashmore_at_neonramp.com> wrote in message
>news:u6vfh4a3t4t3eb_at_corp.supernews.com...
>> Your database may be having trouble completeing checkpoints.
>>
>> Determine how often it is checkpointing.
>> You can set init.ora parameter log_checkpoint_to_alert
>> to true.
>>
>> Also make sure you are not getting a bunch of checkpoint not complete
>errors
>> in your alert.log.
>>
>> If you are doing excessive checkpoints you may try
>> increasing the size of log_buffer.
>>
>> You may also try using multiple db_writers or some io_slaves.
>>
>> Good luck!
>>
>> Stephen C. Ashmore
>> Brainbench MVP for Oracle Administration
>> http://www.brainbench.com
>>
>>
>> "Alexander Müller" <it_mueller_at_web.de> wrote in message
>> news:a4nojc$cmo$1_at_news.online.de...
>> > Hi,
>> > i'm adding 3 datafiles and 3 idx-datafiles in two tablespaces, and then
>we
>> > reorganize parts of the tablespace in this datafiles. After reorganize
>the
>> > database was with the factor 10 slower.
>> >
>> > we bring the sga after this action up and we move the files to another
>> > disks. But it would be not better. At time we have in both tablespaces
>27
>> GB
>> > / Oracle 7.3.4.5.0 / HP-UX 10.20
>> >
>> >
>> >
>> >
>> > --
>> > best regards
>> >
>> > Alexander Müller
>> >
>> >
>>
>>
>

[Additions and Corrections Always Welcome.] Best Regards,
Ganesh R Received on Wed Feb 20 2002 - 02:47:02 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US