Analyzing the performance improvements achieved from moving a volume

You can use Unified Manager to investigate the impact of a volume move operation on the latency (response time) of other volumes on the cluster. Moving a high performing volume to a less busy aggregate or an aggregate with flash storage enabled allows the volume to perform more efficiently.

Before you begin

About this task

Unified Manager identifies when a volume moves between aggregates. It can detect when the volume move is occurring, completed, or failed. The Performance/Volume Details page displays a change event icon for each state of the volume move, which helps you track when a move operation occurred and helps you determine whether it might have contributed to a performance event.

If you are viewing the Event details page, you can click the name of a volume to go directly to the Performance/Volume Details page.

Steps

  1. In the Search bar, type the name of the volume.
  2. Click the name of the volume.
    The volume is displayed on the Performance/Volume Details page.
  3. In the Historic data chart, adjust the sliders to show activity from the previous work week.
  4. Analyze the Latency chart and the IOPS chart to see how the volume performed over the last few days.
    Assume that you notice a consistent pattern of very high average response times of over 42 milliseconds per operation (ms/op), with performance events, each day of the week and decide to move the volume to a less busy aggregate to improve performance. Using OnCommand System Manager, you can move the volume to an aggregate with Flash Pool enabled for an increased performance boost. Approximately an hour after the volume move has been completed, you can return to Unified Manager to confirm that the move operation was completed successfully and that the latency has improved.
  5. If the Performance/Volume Details page is not displayed, search for the volume you want to view.
  6. On the Historic data chart, click 1d to view the activity from the last one day, a few hours since the volume move was completed.
    At the bottom of the page, in the Events time line, a change event icon (Change event icon) is displayed to indicate the time that the volume move operation was completed. A black, vertical line is also displayed from the change event icon to the Latency chart.
  7. Point your cursor to the change event icon to view details about the event in the Events List.
    Because the volume moved to an aggregate with Flash Pool enabled, you can see the change in read and write I/O to cache.
  8. On the Break down data by menu, under MBps, select Cache hit ratio.
    The Cache hit ratio chart displays statistics about the reads and writes to cache.

    The volume successfully moved to a less busy aggregate and the change event is highlighted in the Events List on the right. The average latency decreased significantly from over 42 ms/op to around 24 ms/op. The current latency is around 1.5 ms/op. In the Cache hit ratio chart, the amount of successful read and write hits to cache is now at 100% because the volume is now on an aggregate with Flash Pool enabled.