Grafana Metrics Drilldown on Grafana Labs

About metrics and telemetry

Fri, 03 Apr 2026 19:43:06 +0000

About metrics and telemetry

Metrics, logs, traces, and profiles form the pillars of observability. Correlating the four pillars of observability helps create a holistic view of your application and infrastructure.

Telemetry signals

Metrics

Metrics provide a high-level picture of a system’s state. They are the foundation of alerts because you can compare them against known thresholds. Alerts run continuously and trigger when values move outside expected ranges. Metrics often provide the first signal that something changed.

Logs

Logs provide an audit trail of activity from individual processes that adds context. Logs are atomic events that describe what is happening in your application’s services. While metrics are quantitative (numeric) and structured, logs are qualitative (textual) and unstructured or semi-structured. Logs provide more detail, but they can also create much higher data volumes. Logs let you know what’s happening to your application.

Traces

Traces add detail to the observability picture by showing what happens at each step in a request path. A trace shows where requests start and finish and how long each step takes to complete. For example, a trace can show how long an HTTP request, a database lookup, or a call to a third-party service takes. This data helps you locate problem areas and understand their impact.

Profiles

Profiles help you understand how your applications utilize compute resources such as CPU time and memory. This helps you identify specific lines of code or functions to optimize and improve performance and efficiency.

Why use metrics?

Metrics tell you how much of something exists, such as how much memory a computer has available. In Grafana, metrics are most useful when you record them repeatedly over time. This lets you compare how changes, such as running a program, affect available system resources.

Metrics like these are stored in a time series database (TSDB), like Prometheus, by recording a metric and pairing that entry with a time stamp. Each TSDB uses a slightly different data model, but all combine these two aspects. Grafana and Grafana Cloud can accept their different metrics formats for visualization.

For example, you might be interested in comparing system I/O performance as the number of users increases during a morning while many users in a company come online to start their work days.

A chart showing resource use over time is an example of a visualization. Comparing time-stamped metrics in visualizations helps you spot system changes as they happen.

Grafana and Grafana Cloud offer a variety of visualizations to suit different use cases. For more information, refer to the Grafana documentation on visualizations.

Use metrics with other telemetry data

If you also have other telemetry data, like logs or traces, you can use the telemetry signals together.

Refer to Use signals together to learn about using metrics with other telemetry data and why correlation matters.

Get started with Grafana Metrics Drilldown

Fri, 03 Apr 2026 19:43:06 +0000

Get started with Grafana Metrics Drilldown

Use Grafana Metrics Drilldown to explore your metrics without writing a PromQL query. You can access Grafana Metrics Drilldown in Grafana Cloud or in self-managed Grafana.

Before you begin

The Metrics Drilldown app is installed in both Grafana Cloud and self-managed Grafana by default. Make sure you have the following prerequisites for your environment:

Requirement	Grafana Cloud	Self-managed Grafana (OSS / Enterprise)
Grafana	A Grafana Cloud account with a Grafana stack	Your own Grafana instance running version 11.6 or later
Data source	A configured Prometheus-compatible metrics data source	A configured Prometheus-compatible metrics data source, with the scrape interval set to match your metrics system’s scrape interval

Access Metrics Drilldown

Access Metrics Drilldown either through the main page in Grafana or through a dashboard.

From the Grafana main page	From a dashboard
1. From the Grafana left-side menu, select Drilldown.	1. Navigate to your dashboard in Grafana.
2. From the list of Drilldown apps, select Metrics.	2. Select a time series panel.
	3. Open the panel menu, and select Open in Grafana Metrics Drilldown.

Note
In some Grafana versions, the dashboard menu action appears under Metrics drilldown.

Use metrics with other telemetry data

If you also have other telemetry data, like logs or traces, you can use the telemetry signals together.

Refer to Use signals together to learn about using metrics with other telemetry data and why correlation matters.
Refer to Telemetry signal workflows to explore workflows across all the Drilldown apps.

Next steps

Now you’re ready to drill down into your metric data. For more information, refer to Drill down your metrics.

Drill down your metrics

Fri, 03 Apr 2026 19:43:06 +0000

Drill down your metrics

Drill down into your metrics to gain insight into your data without writing a query. First, filter the metrics that Metrics Drilldown surfaces. Then, select a metric to perform in-depth analysis.

Filter metrics in Metrics Drilldown

To begin drilling down your data, filter the metrics that appear in Metrics Drilldown.

Navigate to the Metrics Drilldown page in Grafana. Refer to Get started with Grafana Metrics Drilldown.
From the Data source dropdown, select a data source to view related metrics. Supported data sources include Prometheus and Prometheus-compatible data sources.
(Optional) Select a label name from the Filter by label values dropdown. Then, follow the prompts to complete your filter criteria.

Note
You can apply multiple filters to your metrics. If you don’t see a metric you expect, refer to Troubleshoot missing metrics.
(Optional) To search for metrics, type keywords in the search bar under Quick search metrics.
(Optional) Select how you want to sort metrics in the app. You can sort metrics alphabetically (A-Z or Z-A), with recently selected metrics first (default), by prevalence in dashboard panel queries, or by prevalence in alerting rules.
Use the time picker to select a date and time range from the dropdown menu, or use an absolute time range.
Select the refresh interval control next to the Refresh icon to set a refresh rate from the dropdown menu. By default, refresh is Off.

Apply advanced filters

Apply advanced filters to further refine the metrics that appear in Metrics Drilldown. Access the following advanced filters from the left-side menu of the Metrics Drilldown app.

Filter type	Description
Rules filters	Filter metrics based on whether they are regular metrics or recording rules.
Prefix filters	Filter metrics based on their name prefix in the Prometheus namespace. Multiple selected prefixes use “OR” logic.
Suffix filters	Filter metrics based on their name suffix. Multiple selected suffixes use “OR” logic.
Recent metrics filters	Filter metrics based on when they started being ingested.
Group by labels	Group metrics by their label values.
Bookmarks	View saved metrics.

Filter logic

Multiple selections for a filter apply “OR” logic, but selections between filters apply “AND” logic. For example, if you select both envoy and kafka as prefix filters, the app shows all metrics with a prefix of either envoy or kafka. Then, if you select count as a suffix filter, the app updates to only show metrics with the count suffix.

Analyze selected metrics

After filtering the metrics in Metrics Drilldown, you can investigate the selected metrics.

From the Metrics Drilldown main page, locate the metric you want to investigate.
From the upper-right corner of the metric panel, click Select.

A detailed view of the metric opens that shows the following details:

A Breakdown tab that shows time series visualizations for each of the label-value pairs for the selected metric. To add a label-value pair to your filters, you can drill down on each label and then select Add to filter.
A Related metrics tab that shows other metrics with similar names and common prefixes. Use it to quickly find metrics that belong to the same area or task without knowing exact metric names.
A Related logs tab that shows how many related log streams match your metric and lets you continue investigating in logs-focused workflows.

View the metric in Explore

Using the panel menu, you can view the metric in Explore or copy the URL.

To open the visualization in Explore, use the metric panel actions and select Explore.

You can share a metric view with others or bookmark it for yourself.

To share, select Copy URL to copy the current page URL to your clipboard. The URL isn’t saved in the app.
To bookmark, select Add bookmark to save your current exploration state, including the data source, filters, selected metric, label breakdowns, and time range.

To revisit a bookmark, open the bookmarks list from the left-side menu. Bookmarks are personal and saved per user. If you want to share a specific state with others, use Copy URL instead.

Troubleshoot missing metrics

Some Prometheus metrics may not appear in Metrics Drilldown, even though you can view them in Explore. This happens in high-cardinality environments when the Prometheus data source enforces its default 40,000-metric limit.

To resolve this issue, choose one of the following options:

Increase the metric limit in your Prometheus data source settings.

Caution
Increasing this limit can affect Metrics Drilldown performance.
Use the name filter to focus on a specific metric. For example: __name__ = vm_http.

Grafana Metrics Drilldown on Grafana Labs

About metrics and telemetry

About metrics and telemetry

Telemetry signals

Metrics

Logs

Traces

Profiles

Why use metrics?

Use metrics with other telemetry data

Get started with Grafana Metrics Drilldown

Get started with Grafana Metrics Drilldown

Before you begin

Access Metrics Drilldown

Use metrics with other telemetry data

Next steps

Drill down your metrics

Drill down your metrics

Filter metrics in Metrics Drilldown

Apply advanced filters

Filter logic

Analyze selected metrics

View the metric in Explore

Share and bookmark metrics

Troubleshoot missing metrics