Understand your metrics usage with the metrics usage report 🔗
To get a detailed breakdown of your metric time series (MTS) creation and usage, you can request a usage report for a specific time interval by contacting your tech support member or your account team. By default, the time interval is 30 days.
You can use the detailed metrics usage report to optimize your usage of custom metrics.
If you are on a usage-based pricing plan, the system counts your metrics as custom metrics. By using the report to find and clean high cardinality metrics, you can better control your costs and query performance.
If you are on a host-based pricing plan with high utilization of custom metrics, you can use the report to lower your custom metrics usage. Learn more about this subscription plan in Infrastructure Monitoring subscription usage (Host and metric plans).
Format of the report 🔗
For each metric in the report, you can see the columns shown in the following table.
Metric identifiers 🔗
Column |
Description |
|---|---|
Metric name |
The name of the metric |
Category type |
The category of the metric, in the format of a number. It only applies to host-based subscription plans. In data points per minute (DPM) subscription plans, all metrics are considered custom metrics. |
The following table has an overview of metric category types. To learn more about metric categories, see Metric categories.
Billing class |
Metrics included |
|---|---|
Custom metrics |
Metrics reported to Splunk Observability Cloud outside of those reported by default, such as host, container, or bundled metrics. Custom metrics might result in increased data ingest costs. |
APM Monitoring MetricSets |
Includes metrics from APM Monitoring MetricSets. See Learn about MetricSets in APM for more information. |
RUM Monitoring MetricSets |
Includes metrics from RUM Monitoring MetricSets. See Filter and troubleshoot with custom tags for more information. |
Default/bundled metrics (Infrastructure) |
|
Default/bundled metrics (APM) |
|
Other metrics |
Internal metrics |
Usage statistics 🔗
Column |
Description |
|---|---|
Detectors |
Number of unique detectors running during the interval for the metric. Detectors created before the interval are also counted. |
Charts |
Number of unique charts containing the metric that users viewed in dashboards during the interval. If users viewed the same chart multiple times, all the views count only once towards the total charts count. |
MTS creation statistics 🔗
Column |
Description |
|---|---|
MTS count |
The number of MTS created for the metric during the interval. |
Common dimensions |
A list of common dimension name combinations for the MTS created along with a number of MTS created with that combination. |
Dimension cardinality |
The approximate cardinality of each dimension for the MTS created for the metric. Error rate is less than 2% when the cardinality is high. |
MTS per token |
The token used to create the MTS and the number of MTS created for the metric using that token.
|
Example MTS |
Three examples of MTS for the metric containing the dimension key-value pairs of the metric. |
Leverage metrics usage report to optimize your metrics volume 🔗
Using the statistics in the metrics usage report, you can gain more visibility into and control over your metrics in the following ways:
Understand the cardinality of your metrics and determine the top cardinality metrics.
Find high cardinality metrics that aren’t frequently used in charts and dashboards. You can optimize your metrics volume by aggregating or dropping these metrics with metrics pipeline management.
Identify dimensions that are the main drivers of high cardinality. You can drop these dimensions and ingest only the aggregated metrics with metrics pipeline management.
Determine which teams are ingesting high cardinality metrics. You can work with those teams to optimize your metrics.
With in-depth insights into your metrics usage and creation, you can make the most use out of metrics pipeline management by aggregating or dropping metrics which are the main cost drivers.
To learn more about metrics pipeline management, see Introduction to metrics pipeline management.