I have found ways to report on latency at the array and storage group level for our VMAX but I would like to see latency at the LUN/device level. We have a few VMs that are not performing at expected levels and we suspect it may be due to I/O latency. These VMs have several disks on multiple datastores. All datastores have a single extent. I would like to find a way to report on read and write latency for each LUN used by these VMs. There must be a way to get this information.
in SPA/Unisphere for VMAX you can drill down to individual device and look at its performance stats (include response times)
Can you provide more details? I have looked in Unisphere for VMAX and have not been able to find the drill down you suggest.
Is there a CLI way to get this information? I rarely use Unisphere for VMAX and I would like something that is offers an easier way to export the data to Excel or a data analysis tool.
When you are in Unisphere in the monitoring section, got to Analyze > Diagnostic > Storage Group. If you double click on the Storage Group you will see the individual volumes in the Storage Group.
I recommend you use Unisphere Peformance Analyzer if you want to analyze the performance of the array. The latest versions have a ton of good data in there that is easily accessible.