SQL Server Performance Forum – Threads Archive

Iometer and SQL Server

I am working to come up with a valid Iometer config file to stress test and compare a NetApp FAS920C storage subsystem against an EMC CX500. We are evaluting the two SANs to determine which one is faster and better for SQL Server. My question is, "How do I determine a fair Iometer config file to run on both systems and what should the workers look like?" I have configured both SANs with two volumes (one for the data file and one for the log file). The volumes have idential spindle counts, drive speeds, etc… I need to come up with workers for the data volume and workers for the log volume. Does anyone have any good suggestions?

This is going to depend on what type of databases you have on SQL Server and what you are using them for. Could you give us a little more information? MeanOldDBA
[email protected] When life gives you a lemon, fire the DBA.

let me tell you ahead of your tests,
if you test random writes first, then sequential read or writes,
the netapp should win the random write, but perhaps not the sequential op,
the netapp has a special method for speeding up random writes (write anywhere?), at the possible expense of subsequent sequential ops. if you are going to SQL Connections in Orlando, stop by to chat on this

I would like to have a chat with you on this Joe. <img src=’/community/emoticons/emotion-1.gif’ alt=’

‘ /> I’ve enjoyed reading your posts, since you’ve actually done a lot of the things you talk about. I’ll be in Orlando, so maybe we can talk for a few minutes. It’s nice to know that someone else has worked with SANs directly related to SQL Server. A lot of the information I found out there ended up having no touch with reality. It was a good read though. MeanOldDBA [email protected] When life gives you a lemon, fire the DBA.

]]>