SQL Server: IOps in 2008 R2 vs 2005

I was given a report by our hosting provider, who manages our SAN, that shows the peak number of IOps on our SAN-attached drives (although not the throughput) has gone way, way up.  In some cases the peak IOps is almost seven times as many IO read/writes for only slightly more MB/sec of data since last year.  The major difference between this year and last is that we upgraded to SQL 2008 R2 from SQL 2005.
Is there any reason to think that SQL 2008R2 might be doing more, but smaller IOps?  Or might be bursting them more rapidly?  I don't have a good explanation for it (and the boss is asking.)  To cover all bases, I have to at least wonder about the veracity of the report he gave me (the report is an analysis of an EMC NAR file, if that means anything to you).   We have more total processor cores on servers connected to it, as well - could that have something to do with it (more threads doing reads and writes)?
cgilsonAsked:
Who is Participating?
 
cgilsonAuthor Commented:
After our hosting provider took another look at it, it turns out that when we created some new LUNS, they created them with a dramatially smaller block allocation size than what had been configured on the server previously. Fortunately, we're going to be moving out of this data center in the next couple months, so we don't really have to worry about .  This explains it!
0
 
mark_gaoCommented:
sql 2008 has a data compression feature
this feature is used by db mirroring  automatically, can reduce io
and the compression feature for backup have a high data compression ratio
0
 
AlokJain0412Commented:
There is problem in up-gradation from 2008 r2 settings and Mistakes in  follow Upgrade adviser


would you recall what process you follow in up-gradation that is effecting you server IOps

Following link will be use full
http://www.sqlmag.com/content1/topic/upgrading-a-sql-server-2005-cluster-to-a-sql-server-2008/catpath/clustering/page/2
0
Ultimate Tool Kit for Technology Solution Provider

Broken down into practical pointers and step-by-step instructions, the IT Service Excellence Tool Kit delivers expert advice for technology solution providers. Get your free copy now.

 
cgilsonAuthor Commented:
Sorry, I don't understand your first statement.  Could you please rephrase it?  
To upgrade the server, we did a completely new install of SQL on a new server, disconnected jkconnected the LUNS where the user databases resided to the new server, and attached them using CREATE DATABASE ... FOR ATTACH statements, created the server-level objects (logins, linked servers, etc) with scripts, then upgraded the databases with sp_dbcmptlevel.
0
 
AlokJain0412Commented:
sorry for typing mistake
i would like to write
"There is problem in up-gradation from sql server 2005 to 2008 r2 settings and Mistakes in  follow Upgrade advise"

this is t link the link which helps you to determine you have installed and configure properly

http://www.techrepublic.com/blog/10things/10-tips-for-upgrading-to-sql-server-2008-r2/2376

http://technet.microsoft.com/en-us/magazine/gg454217.aspx

Next

If i catch correctly your question means that you are facing High I/O with  sql 2K8 r2 server and slow performance?

then specify it with complete specification configuration of your all systems  Os, HD,





0
 
cgilsonAuthor Commented:
We're not actually haveing a performance problem, my boss is just asking why the IO is so much higher, and I have no good explanation for it - That's part of the reason that I think the report might be wrong (or at least, it's capturing wrong or incomplete data).  I was wondering if SQL Server 2008 R2 somehow did more, smaller-sized reads and writes than SQL 2005.  I haven't been able to find anything that says that is the case.  Also, since the upgrade advisor didn't say we really needed to change anything, and we didn't actually perform an in-place upgrade (instead we moved the databases to an entirely new server running the newer version of SQL), I don't think we did anything "wrong" per se, in the upgrade.
 I think I am going to tell them to make sure that the report is accurate, or to find out if anything else changed on that SAN.
0
 
AlokJain0412Commented:
hi
in my opinion there is something wrong  with configuration,hardware software memory settings,OS or version problem,
you please scan following thread  
if would useful
http://www.sqlteam.com/forums/topic.asp?TOPIC_ID=154938

this link
Tuning your SAN: Too much HBA Queue Depth?

http://sqlblogcasts.com/blogs/christian/archive/2009/01/12/tuning-your-san-too-much-hba-queue-depth.aspx
0
 
cgilsonAuthor Commented:
The solution was from something I had initiated before recieving any helpful comments.
I gave myself a "B" because I didn't make sure that the drives were configured correctly when I accepted them from the hosting provider!
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.