When you specify a Sensitivity level, the management console automatically generates a new threshold value for a metric each night at GMT midnight, using percentile statistics from the last 30 days to determine an appropriate setting. The management console generates a separate set of threshold values for the users who access an application from each client network.
As a result, the management console does not apply the same threshold value to users on high-latency remote links as it does to users on a local LAN, and a threshold value represents historically extreme performance. For example, consider an application server located at a data center in Miami, with users who access the application locally and from a remote site in Munich, Germany.
If you use Sensitivity levels to set performance thresholds for this application, the management console generates a separate set of threshold values for the users who access the application from each client network. Specifying Sensitivity levels also means that you are not required to define the same application twice, such as Exchange to Munich and Exchange to Miami, to set different thresholds for application performance across remote locations.
When the sensitivity of a threshold is raised, the threshold value is lowered because the intent is to receive more incidents based on performance levels. The management console sensitivity threshold is similar to pain sensitivity. People who are highly sensitive have a low threshold for pain, and will cry out more often. People who are less sensitive can handle more pain and are much less likely to cry out.
Sensitivity can be adjusted from a value of 0 (insensitive) to 200 (very sensitive):
The Sensitivity level does not change when the management console recalculates threshold values. To view today’s threshold value for a Sensitivity level, use the Sensitivity calculator.
|
Copyright © 2014 CA Technologies.
All rights reserved.
|
|