Regrading Performance issues for servers

Also could any one let me know what are the minimum values these things need to be .....

Are you monitoring response times on the hosts? This is probably the best metric to determine storage health.

Otherwise getting an EMC Symmetrix SPEED guru involved would be the quickest way to determine if the Symm has an issue or not.

What kind of host OS is this? If Windows host, turn on data collection for perfmon (LogicalDisks & PhysicalDisks), Check read and write service times. For Unix based hosts check iostat and/or "sar -d" data to see what the service time and IO load looks like. As Quincy56 mentioned, if the service times are high (>usual (good day) or >25ms) then definitely plan to spend some looking at the DMX. Also check for queued IOs.
If you have the iostat/sar/perfmon data, I can help by using some of the tools i have for this.

I think the first and most important question is to the LOB reporting the problem. They need to let you know what evidence they have that the storage is the root of the problem. What are they actually seeing.

Then the investigation should start at the server. If the LOB have valid complaints and the server investigation supports the belief that the Storage is involved, the server investigation results should be able to direct you to the most appropriate things to check on the DMX.

Hello Vajinap

There are certain paramter which help you to figure out the DMX performance.

And if you have symmetrix performance analyzer tool then it will be a great help to analyze the dmx performance.

here are the thinhs you need to check :

Check Cache Performance:

Run the System à System Cache Utilization ViewGraph. You will see

What is the average cache hit percentage for this workload?

What is the worst cache hit ratio observed in this data?

Was the hit ratio inflated by write activity?



Check Balance Across the Front End:

Run the Dir-Fibre>All Fibre Directors-IO Rates ViewGraph. You will see

Is the activity constant over time?

Is the activity balanced across all the channels?



Check Balance Across the Back End:

Run the Dir-DA>All Disk Directors > IOs per Sec ViewGraph. You will see

Is the activity balanced across the back end?



Check Balance Between Front End and Back End:

Run the System>All Directors ¿ IOs per Sec ViewGraph.

Is the front end more active than the back end?



Check Logical Volume Balance:

Run the Devices>Average IOs per Sec ViewGraph.

Are the average IOs per second spread uniformly?

Does there appear to be any significant striping?

There are a number of steps that EMC goes through when troubleshooting performance issues.

Firstly the host configuration

Are the latest HBA drivers and firmware and OS patches recommended by EMC applied along with the recommended multipathing software

For OSes that rely on partition offsets then are these correct for optimal performance

Next the switches

How many switches are involved in the communication to the storage

are there any issues reported on the ports facing the HBAs or the ports facing the storage

are there any issues reported on any ISLs in the fabric (even if they do not appear to be on the route between host and storage)

Is switch firmware at recommended / supported levels

Finally the storage

We would possibly need to engage a local Performance expert to analyze the behaviour seen on the storage - gathering stp data for DMX/V-Max or NAR files and SPCollects for Clariion

This last step should only be carried out once everything else has been eliminated

Obviously this would need to be done in the context of a Service Request.