Please start any new threads on our new site at https://forums.sqlteam.com. We've got lots of great SQL Server experts to answer whatever question you can come up with.

 All Forums
 SQL Server 2005 Forums
 SQL Server Administration (2005)
 Disabling Write Caching

Author  Topic 

Michael Valentine Jones
Yak DBA Kernel (pronounced Colonel)

7020 Posts

Posted - 2008-06-30 : 14:26:24
I got this is a document from a vendor:
"For maximum performance, Microsoft recommends disabling write caching on all SQL Server disk drives."

I have heard of disabling write caching for data safety, but never for performance reasons. Anyone ever hear of disabling write caching for SQL Server performance? It seems counter-intuitive to me, and I would not be surprised if this vendor was just dead wrong again.






CODO ERGO SUM

sodeep
Master Smack Fu Yak Hacker

7174 Posts

Posted - 2008-06-30 : 14:50:58

Check this out:

http://www.sql-server-performance.com//tips/hardware_tuning_p4.aspx
Go to Top of Page

maninder
Posting Yak Master

100 Posts

Posted - 2008-06-30 : 15:15:16
We have physical drive write cache disabled, which is the default setting on our HP smart array controllers. As per HP, the physical drive write cache should be disabled as it is not battery-backed.

Maninder
Go to Top of Page

Michael Valentine Jones
Yak DBA Kernel (pronounced Colonel)

7020 Posts

Posted - 2008-06-30 : 15:20:22
I understand that you might want to disable write caching for safety, but has anyone ever heard of disabling it for performance reasons?

This seems to say the opposite:
http://support.microsoft.com/kb/86903




CODO ERGO SUM
Go to Top of Page

maninder
Posting Yak Master

100 Posts

Posted - 2008-06-30 : 16:19:54
Check this out http://support.microsoft.com/kb/234656 and http://sqlserver-qa.net/blogs/perftune/archive/2007/05/22/battery-backup-write-cache-for-controllers-should-it-be-disabled.aspx

Maninder
Go to Top of Page

Haywood
Posting Yak Master

221 Posts

Posted - 2008-06-30 : 18:11:04
No, not for performance reasons...

I suggest you read (and perform your own tests) Linchi's recent articles regarding disk controller cache:


http://sqlblog.com/blogs/linchi_shea/archive/2008/03/24/performance-impact-of-controller-read-cache-large-sequential-i-os.aspx
http://sqlblog.com/blogs/linchi_shea/archive/2008/03/27/performance-impact-of-controller-cache-sql-server-large-operations.aspx
http://sqlblog.com/blogs/linchi_shea/archive/2008/03/29/performance-impact-of-controller-cache-sql-server-oltp-workloads.aspx
http://sqlblog.com/blogs/linchi_shea/archive/2008/04/01/performance-impact-of-controller-cache-sql-server-read-only-workloads.aspx




Go to Top of Page
   

- Advertisement -