Re: filesystemio_options setting
Date: Fri, 16 May 2008 10:44:31 +0300
Normally you use 3 caches for blocks:
1) Buffer pool 2) File system cache 3) Storage cache.
You access the storage system only if the block is not found in the first 2. When you access the storage system you might find the block in the storage cache.
By bypassing the file system cache you need to access the storage system
The storage system needs to keep in it's cache the blocks previously held in the file system cache, effectively decreasing the hit ration for the cache.
This increase the number of physical I/Os and thats leads to degradation.
Depending on your workload this can negate the performence benefit you get from doing direct I/O.
You can check it easily but turning the setall again and asking the EMC guys
to monitor the storage cache efficiency and number of physical I/O's
for your disks.
John Smith wrote:
> Hi. We are running oracle 10.2.0.3 <http://10.2.0.3> on our test
> evironment in suse 9. Emc symmetrix storage. I was going over the
> settings and discovered that filesystemio_options was not set (no
> value given). I changed it to setall per the Oracle documentation for
> Linux. Shortly after that, our queries suddenly started running
> longer, and the culprit appears to be disk IO.
> I understood that this setting could not cause a problem, and all the
> Oracle documentation I could find said to set this value to setall in
> 10.2.0.3 <http://10.2.0.3>. I unset the value and the problem went
> away. Can someone shed some light on this?