Analyze your dimensions with Usage Analytics
Use the Dimensions tab in Usage Analytics to explore how contextual dimensions affect your metric footprint.
View and explore your dimensional usage
The Dimensions tab in Usage Analytics provides interactive charts and detailed tables that reveal how contextual dimensions contribute to your overall metric footprint. It’s designed to help you assess dimensional cardinality, distribution, and usage across your environment.
A summary card displays the total number of dimensions and their unique values, offering a quick snapshot of dimensional richness. A trend chart shows how the number of dimensional values changes over time, helping you spot growth patterns or anomalies.
The core of the tab is the dimensions table, which lists each dimension alongside key usage indicators:
- Presence in detectors, dashboards, charts, and API queries
- Number of unique values and their share of total dimensional values
- Contribution to metric time series (MTS) volume
- Association with specific metrics
You can sort and filter the table to focus on dimensions with high cardinality, broad usage, or optimization potential. Clicking a dimension opens a detailed profile.
Understand dimensions usage with the dimensions table
The dimension usage table displays the following fields:
Field | Description |
|---|---|
Dimension name | The name of the dimension. |
Utilization | Indicates how a dimension is used across your environment. Each dimension may appear in multiple Observability Cloud components, but in the Utilization column, it is associated with the highest rank in which it is used. This does not mean the dimension is used only in that one component. For more details, see the following list:
Example:If a dimension is referenced in both inactive charts and dashboard filters, Usage Analytics assigns the dimension to the highest applicable utilization rank. In this case, the dimension would be marked as used in dashboard filters. |
| Detectors | Indicates the number of detectors in which this dimension is used, as a part of the metric. |
| Dashboard filters | Number of dashboards where this dimension is used as a top filter. This usage is independent of any metric context. |
| Active chart | Indicates the number of active charts in which this dimension is used, as a part of the metric. An active chart is one that has been viewed in the given time period. |
| API queries | Indicates the number of API queries that reference this dimension. |
| Inactive charts | Indicates the number of inactive charts in which this dimension is used, as a part of the metric. An inactive chart is one that hasn't been viewed in the given time period. |
| Unique values | The average number of unique dimensional values per hour. |
| Percentage of total values | The average number of dimensional values (for the given dimension) over the total number of all dimension values per hour. |
| Metrics with dimension | The average number of metrics with the dimension per hour. |
| MTS with dimension | The average number of MTS with the dimension per hour. |
| Percentage of total MTS | The average number of MTS with the given dimension over the total number of all MTS per hour. |
Show hidden utilization columns
The dimensions table includes several detailed utilization columns: Detectors, Dashboard filters, Active charts, API queries, and Inactive charts which are hidden by default. These columns are designed to provide a quick, high-level view of how often each dimension is used within each component, helping decide whether to explore a dimension in more detail and identify underutilized or high-impact dimensions.
To display them:
Select Settings in the usage analytics table.
Under Show/Hide Columns, select each utilization column to show in the table.
Filtering dimensions by utilization
You can filter dimensions in the usage analytics table to focus on specific utilization levels, such as unused dimensions, actively referenced dimensions, or those tied to detectors and API queries.
To apply a utilization filter:
- Select the filter box labeled Utilization: Any at the top of the page.
From the dropdown menu, select the desired utilization level (e.g., Unused, R3 – Active charts, R4 – Detectors).
Select Run search to apply the filter.
The table and charts will update to show only dimensions matching the selected criteria. To clear the filter and return to the full view, select Reset.
Dimensions profile: contents and structure
Usage Analytics includes dimension profiles for each of your dimensions. These profiles provide a comprehensive view of how a dimension is used across your environment and how it contributes to your metric footprint.
Within the profile view, you can filter the data by Billing class and Time period to narrow down the scope of analysis.
To access a dimension profile, select any dimension directly from the dimensions table.
Each dimension profile includes the following tables:
Table | Description | Notes |
|---|---|---|
Metrics | This tab provides the primary interface for managing dimensions. Dimensions cannot be removed globally; they can only be removed within the context of a specific metric. By selecting the Settings icon next to a metric and choosing Aggregate dimension, you'll be redirected to Metric Pipeline Manager (MPM), where you can remove the dimension from that metric. The Dimension utilization column in this tab shows how the selected dimension is used specifically within each listed metric — for example, whether it's used in detectors, charts, or queries. This contextual view helps identify optimization opportunities at the metric level. | Displays up to 10 000 results. |
Dashboard filters | Lists dashboards where the dimension is used as a filter, along with the user who last updated each dashboard and the time of the update. | Displays up to 10 000 results. |
Charts | Displays dashboards containing charts that reference the dimension, including the user who last updated each chart and the timestamp of the update. | Displays up to 10 000 results. |
Detectors | Shows detectors that reference the dimension, along with the user who last updated each detector and the time of the update. | Displays up to 10000 results. |
| Sample values | Shows a selection of values for the dimension. | Displays up to 10 randomly selected values per dimension. |
Manage and reduce your metric usage
This section provides practical strategies for identifying dimensions that can be aggregated, archived, or dropped to help reduce your overall metric time series (MTS) volume and simplify your environment.
Archive or drop unused dimensions
Using the dimensions table, you can find dimensions that aren't actively used in detectors, charts, dashboards, or API queries. If you have unused dimensions, consider dropping or aggregating them to reduce your usage footprint and simplify your environment.
Dimensions that are no longer actively used can be discarded to reduce billing impact. If you don't plan to use certain dimensions in the future, consider dropping them entirely to free up usage capacity. You can reintroduce them later if needed by updating your metric definitions.
Aggregate dimensions with low utility
Dimensions with low utility may be strong candidates for metrics aggregation rules, helping reduce the overall number of metric time series.
By understanding how dimensions are utilized, you can make informed decisions about whether to retain or discard them. Selecting specific dimensions to keep allows you to aggregate data points into a new metric with fewer dimensions, resulting in a focused view of the dimensions that matter most.
To evaluate whether aggregation is appropriate:
- Select the dimension from the dimensions table to open its profile.
Identify metrics where the dimension is not actively used — these can be safely excluded.
- For metrics where the dimension is in use,review the Detectors and Charts tabs to check where the dimension is used.
Evaluate the dimension's relevance in each context. If it's not essential, consider creating an aggregation rule to remove it from the metric definition.
To learn more, see Use aggregation rules to control your data volume.
Consider rerouting part of your metric to the archived data route
If you're unsure whether to drop or keep a dimension, consider archiving the entire metric and creating an exception rule. This rule will allow metrics with dimensions you're confident about to remain in the real-time data route.
To learn more, see Use routing exception rules to route a specific MTS or restore archived data.