White Papers
Table Of Contents
- 1 Introduction
- 2 Generic Recommendations
- 3 Test Environment and Methodology
- 4 Scalability and Performance tests for Power Manager 2.0
- 5 Addition of devices and groups in Power Manager
- 6 Metric Collection
- 7 Power Policy (Power Capping)
- 8 Emergency Power Reduction
- 9 Alert Thresholds
- 10 Importing Physical Groups and Device Association from a CSV file
- 11 Plugin Actions (Install, Disable, Enable, Uninstall)
- 12 Longevity Test Results
- 13 Troubleshooting
- 14 Conclusion
- 15 Technical Support and Resources.
Alert Thresholds
23 Benchmark the Performance, Reliability and Scalability of Dell EMC OpenManage Enterprise Power Manager
2.0 in Data Center Environment
9 Alert Thresholds
Power Manager leverages the Events and Alerts feature in OpenManage Enterprise and supports different
alerts such as Threshold violation alerts, Group membership change alerts, Policy Violation alerts and so on.
Power and Temperature Threshold violation alerts are used to notify the administrator when the power
consumption and temperature has exceeded certain configured values so that the administrator can take
immediate actions.
9.1 Configuration Details and Generic Recommendations
9.1.1 Configuration Details
Violation of Power and Temperature thresholds’ and corresponding event generation has been tested
simultaneously on 250 devices in a configuration where 8000 devices are monitored in Power Manager with
no alert actions configured.
9.1.2 Recommendations
After analyzing the test results, following are the recommendations for Threshold Alerting functionality of
Power Manager.
Note: The benchmarks mentioned in this document for Alert Thresholds are common for OpenManage
Enterprise console.
• Reception of alerts is supported with 5,000 alerts per minute for 60 minutes without any drop when no
alert actions are configured. This implies that the appliance receives 85 alerts per second
continuously for 1 hour without any alert drop.
• With one alert action (email or trap forwarding) configured and mapped with 8,000 discovered
devices, reception of alerts is supported up to 85 alerts per second for a duration of 60 minutes with a
possibility of 20 percent alert drop. Restarting OME services help to receive a greater number of
alerts.
• With all the alert actions configured; reception of alerts is supported till 9 alerts per second with a
possibility of 25% alert drop, if continued for long duration. Restarting OME services help to receive a
greater number of alerts.
• Key factors determining alert reception in case of configured alert actions are:
- Number of devices selected.
- Time interval between alert reception.
- Number of alert actions configured.
The graphs in the following test results represent the CPU and Memory utilization of event processing service
captured during alert reception.