Want to protect your cyber security and still get fast solutions? Ask a secure question today.Go Premium

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 457
  • Last Modified:

SAN Disk I/O metrics

Hi,
I'm currently building a new Windows 2008 R2/SQL 2K8 Failover Cluster and waiting for SAN disks provisioning from the storage team. I am also planning to run a disk I/O measurement tool and save the result as a baseline metrics for future references.

One tool comes in mind is SolarWind SAN Monitor for EMC Clariion.

Are there any other good free tools that I can use? And also, what are the key elements that are essential to disk performance and I should incorporate into my test? Almost new to this area.

Thanks, Sharad
0
sharad_rai
Asked:
sharad_rai
  • 5
  • 5
1 Solution
 
Racim BOUDJAKDJIDatabase Architect - Dba - Data ScientistCommented:
I recommend the following tool for measuring both IOPS (in KIops) and throughput (in MB/sec).  In itself performance monitor will give you a good indication whether the system presents IO contention by looking at LogicalDisk/PhysicalDisk:Avg Sec per Write/Read
0
 
sharad_raiAuthor Commented:
Hi Racino,
Is this part of the Windows Performance monitoring tool?
0
 
sharad_raiAuthor Commented:
Please ignore my last comment.

I just need help deciphering the graph. The metrics uses the default scale of 1,000, what is this scale and how to interpret and relate it to my reading?

And what are those units on Y axis, it starts at 0 and goes up to 100. X is time and moves every second... that I figured

thanks a million,

Sharad
0
Problems using Powershell and Active Directory?

Managing Active Directory does not always have to be complicated.  If you are spending more time trying instead of doing, then it's time to look at something else. For nearly 20 years, AD admins around the world have used one tool for day-to-day AD management: Hyena. Discover why

 
Racim BOUDJAKDJIDatabase Architect - Dba - Data ScientistCommented:
By default the Avg Sec per Write is expressed in ms.  For example, under a scale of 1000,, a contention of 10 on the graph means each Write takes in average 10 ms to run.
0
 
Racim BOUDJAKDJIDatabase Architect - Dba - Data ScientistCommented:
You can also use another tool called IOMeter.
0
 
sharad_raiAuthor Commented:
Hi Racimo,
I installed IOMeter on a test server and it looked pretty good to me. It does however prepare the disks before running the monitoring on them, hope it is safe to do on production disks right?

I found it better than SolarWinds SAN Monitoring as SW works only with specific EMC storage models.

thanks,
0
 
Racim BOUDJAKDJIDatabase Architect - Dba - Data ScientistCommented:
IOMeter is the reference tool for serious OS level IO testing no matter what is the storage layer behind.  It is safe to run but, as any stress tool, it has to put some heat on the partition where the data is stored and may slow down things a little: you can stop it any time though.  Use it, based on application IO profile, to evaluate how much throughput you are getting.
0
 
sharad_raiAuthor Commented:
Excellent, thanks!
I do like this product though...
0
 
sharad_raiAuthor Commented:
Exactly what I was looking for...
0
 
Racim BOUDJAKDJIDatabase Architect - Dba - Data ScientistCommented:
Glad it helped. :)
0

Featured Post

Has Powershell sent you back into the Stone Age?

If managing Active Directory using Windows Powershell® is making you feel like you stepped back in time, you are not alone.  For nearly 20 years, AD admins around the world have used one tool for day-to-day AD management: Hyena. Discover why.

  • 5
  • 5
Tackle projects and never again get stuck behind a technical roadblock.
Join Now