Documentation Grafana Cloud Adaptive Telemetry Adaptive Metrics Monitor

Monitor and validate results

The Overview page is the landing page for Adaptive Metrics. It provides a summary of pending recommendations, a graph showing time series reduction, and quick actions for reviewing and applying recommendations.

Overview page

When you navigate to Adaptive Telemetry > Adaptive Metrics, you land on the Overview page.

Recommendations summary

When there are pending recommendations, the Overview page displays a summary bar at the top of the page. This bar shows:

• The number of actionable recommendations for the selected segment.
• The estimated series impact if you apply all recommendations—shown as either a series decrease or series increase.
• A Review button that navigates to the Rules page where you can inspect individual rules.
• An Apply all recommendations button to apply every pending rule at once.

Use the Segment dropdown to filter rules by segment. For more information about segments, refer to Configure segmentation.

Enable auto-apply from the Overview page

If auto-apply is available for your instance, an Enable auto-apply button appears next to the segment selector. Click this button to select which segments should have auto-apply enabled. For more information, refer to Configure auto-apply.

After you apply all recommendations, a prompt encourages you to enable auto-apply for the current segment so that future recommendations are applied automatically. You can dismiss this prompt or choose not to show it again.

Time series reduction graph

After reviewing and applying recommendations, or using other customization options, you can verify that your setup is delivering the results you expect.

The Time series reduced by applied rules graph shows the difference between raw series and aggregated series over time. This graph appears after you have applied at least one rule.

1. Navigate to Adaptive Telemetry > Adaptive Metrics.
2. From the Overview page, select the segment you want to review from the Segment dropdown.
3. Use the time range picker to include when you first applied rules, and the period after the changes.
4. Check the graph to confirm that there was a reduction after applying the rules.
5. Click Explore to view more detail on the underlying metric or query.

Estimate potential savings

To estimate the potential savings from applying all pending recommendations, query the grafanacloud_instance_recommendations_estimated_savings_series metric from the grafanacloud-usage data source.

This metric represents the expected change in active series if you apply all pending recommendations. In other words, it answers “what change do we expect if you press Apply all recommendations?”

Key points about this metric:

• It includes all recommendation types: Add, Keep, Update, and Remove.
• The value represents the net change in active series, combining all recommendation types.
• The value can be negative if the recommendations include removing rules or modifying rules to drop fewer labels that would result in an increase in total active series.

To view the estimated savings:

promql Copy

sum(grafanacloud_instance_recommendations_estimated_savings_series)

Alternatively, you can see the total impact by clicking Apply all recommendations in the UI. The confirmation dialog box displays the expected series change before you confirm.