RE: DBWR - How Many is Too Many?
- From: Carel-Jan Engel <careljan@xxxxxxxxxx>
- To: mwf@xxxxxxxx
- Date: Thu, 28 Feb 2008 16:46:54 +0100
On Thu, 2008-02-28 at 09:13 -0500, Mark W. Farnham wrote:
>
>
> Finally, coming to the end of my off the top of my head list of things
> I’d give a quick check, is the wait i/o on your system actually from
> the dbwriters? From the database you’ve also got the log writer and
> the archiver and you didn’t mention any data guard issues, but you’ve
> probably also got sys admins practicing backup and recovery using
> spare space on your disk farm. It is a NEW SAN, right?
It seems you have put your control file / online redo log files also on
CIO/JFS2 devices? From what I have understood about this matter, OLRF
should _not_ be on CIO mounted devices, because LGWR will write small
chunks (typically 512bytes) and the Filesystem has bigger blocks (e.g.
8K). When on CIO/DIO, every write by LGWR forces the OS to read the 8K
block, fit the 512 bytes at the correct position, and write it back.
Some FS caching there is quite helpfil then: Just fit the 512 bytes
where it belongs, and write the block, keeping it in cache. The same for
control files, and archived redo log files. Only data files should be on
CIO filesystems. IIRC.
Best regards,
Carel-Jan Engel
===
If you think education is expensive, try ignorance. (Derek Bok)
===
Other related posts:
