The Time To Threshold Event Rule
INLINEConfigure the Time To Threshold alarm for QoS-enabled probes to send an alarm when a QoS metric is predicted to reach a user-defined value within a configured time period. The Prediction Value has the same implicit units as the QoS metric you are configuring.
Configure the Time To Threshold alarm for QoS-enabled probes to send an alarm when a QoS metric is predicted to reach a user-defined value within a configured time period. The Prediction Value has the same implicit units as the QoS metric you are configuring.
To use Time To Threshold, you must have the following probe versions installed at the secondary hub level:
- baseline_engine 2.34 or later
- Probe Provisioning Manager (PPM) 2.38 or later
- prediction_engine 1.01 or later
Time To Threshold is an event violation rule that sends an alarm if a QoS metric is predicted to reach a set value within a user-defined time period. Setting a Time To Threshold alarm for any of the QoS-enabled probes allows the prediction_engine probe to gather trending information used to calculate when a particular event might occur. The Time To Threshold settings are configured using the Admin Console.
At any time, the condition that triggers the Time To Threshold alarm can change (for example, files were deleted to free up space) or the Time To Threshold alarm settings can be reconfigured.
Time To Threshold Workflow
The following diagram shows the process of how QoS data from a QoS-enabled probe is used to generate a predictive alarm.
When you configure the Time To Threshold settings on a QoS-enabled probe, the following actions occur:
- data_engine stores raw QoS metric data from the QoS-enabled probes
- prediction_engine computes trend information for QoS metrics that have Time To Threshold configured
- baseline_engine probe evaluates QoS trend metrics from probes against static threshold definitions
- baseline_engine probe generates threshold violation messages when the Time To Threshold trend timeframe and prediction thresholds are crossed
How a Time To Threshold Alarm is Generated
The process of generating a predictive Time To Threshold alarm includes the following actions:
- The QOS-enabled probe sends raw data to the data_engine probe.
- The data_engine probe stores the raw data in the S_QOS_DATA and S_QOS_DEFINITION tables.
- The prediction_engine keeps samples collected during the hour.
- At five minutes past the hour, the prediction_engine probe calculates a linear regression line based on an hourly median for all QoS metrics with the Time To Threshold alarm configured.
- The prediction_engine sends a new QoS metric, as its sample value, containing the number of hours until the configured QoS reaches the prediction value.
- The baseline_engine probe evaluates these QoS metrics from the prediction_engine against the configured static threshold definitions (time threshold, time units, and prediction value).
- The baseline_engine probe sends an alarm of the configured severity if this new QoS metric is less than or equal to the time threshold (time threshold and time units) configured for each Time To Threshold alarm.
- An alarm is sent each hour for the QoS metric as long as the calculated time prediction is less than or equal to the configured time threshold (time threshold and time units).A Time To Threshold Alarm is no longer sent when the calculated values indicate that the prediction has been met or that the prediction is beyond the configured Time Threshold.
For more Information: