Register Metrics Page

Use the Register Metrics Page to:

Register metrics to be included in baselines of this target.
Deregister metrics currently registered for this target.
Test the quality of (registered or unregistered) metric data for use in Adaptive Thresholds or Metric Normalization baseline features.

Usage Tips

Registering Metrics: Adaptive Thresholds and Baseline Normalized View require metric baseline statistics generated from raw metric data. These statistics are computed only for registered metrics. Therefore, register any metrics desired for use in Metric Normalizations or Adaptive Thresholds for any baseline of the target.

Deregistering Metrics: Deregister metrics when they are no longer needed for metric baseline features. If a deregistered metric is re-registered, only the Moving Window baseline will recompute statistics for that metric and enable Adaptive Thresholds or Metric Normalization features. Static baselines permanently lose metric statistics when the metric is deregistered.

Note: Deregistering metrics will remove generated statistics and Adaptive Threshold settings for those metrics on all baselines.

Testing Metrics
- Click Test All within the Unregistered Baseline Metrics section to evaluate suitability of unregistered metrics for baseline features. The Results column will contain one of three icons after the test is completed. A critical icon indicates there is insufficient data to produce the most useful significance level statistics. A warning icon indicates there is sufficient quantity of data, but the data values tested indicate that this may be a poor candidate for Adaptive Thresholds. A green check icon indicates the metric is reasonably well suited for significance level based baseline functionality and should be considered for registration.
- Click Test All within the Registered Baseline Metrics section to evaluate currently registered metrics. Metrics that consistently test as poorly suited for Adaptive Thresholds and Metric Normalizations should be considered for deregistration to conserve compute resources as statistical computations may not be adding value. The Test Data Window Size drop-down can be used to evaluate a wider window of data. This page may take proportionally longer to return test results for longer window sizes.

Related Topics

About Metric Baselines