Identifying bully workloads involved in a performance event

In Unified Manager, you can identify which workloads have the highest deviation in usage for a cluster component in contention. Identifying these workloads helps you understand why certain volumes on the cluster have slow response times (latency).

Before you begin

About this task

The Dynamic Threshold Event Details page displays a list of the user-defined and system-defined workloads ranked by the highest usage of the component or most impacted by the event. The values are based on the peaks that Unified Manager identified when it detected and last analyzed the event.

Steps

  1. Display the Dynamic Threshold Event Details page to view information about the event.
  2. In the Workload Details table, select Bullies - Peak Deviation in Activity or Bullies - Peak Deviation in Utilization.
    Note: By default, the table is sorted on Victims - Peak Deviation in Latency. When the Workload details table is sorted by peak deviation in usage, workloads with low deviation in usage are not displayed in the table.
    The workloads with the highest deviation in usage are displayed at the top of the table.
  3. Click the name of a volume workload to view detailed information about its current and historical performance activity on the Performance/Volume Details page.