preview
We're still working on this feature, but we'd love for you to try it out!
This feature is currently provided as part of a preview program pursuant to our pre-release policies.
Cardinality refers to the number of unique elements in a set, specifically the unique attribute combinations observed for a given metric within a one-day period. This data resets daily at UTC midnight. As cardinality increases with more unique combinations, it can impact data limits. Understanding and managing cardinality is crucial to optimize data usage and preventing limit issues.
As a New Relic administrator, you oversee data ingestion to ensure it stays within the specified limit. Cardinality Management offers graphs and tables to detail your data usage, allowing you to take action if you're approaching your limits.
To view the Cardinality Management UI, go to one.newrelic.com > (click your user name) > Administration > Cardinality Management.
Account cardinality breakdown
If your organization has multiple accounts, you can view a breakdown of cardinality for each account. To switch accounts, use the account picker in the top right corner. If you have access to the account, you can see the account cardinality information on the UI.
Usage over time
On the landing page of Cardinality Management, you can view the Usage over time chart. The chart displays the account cardinality usage compared the limit of that account.
팁
If the cardinality usage value remains unchanged, then a new data point isn't reported. So, you may see an empty data point. This isn't data loss.
To adjust the timeseries, use the time picker in the top right corner. This change only affects the timeseries in the chart and does't impact the Metrics table following it.
Metrics table
Check the top 20 metrics with the highest cardinality in the selected account for the day. View the breakdown of account cardinality in the Metrics table, organized by metric name.
All metrics have a default cardinality limit of 100,000. Based on your requirement, you can adjust the per-metric limits.
Metric cardinality breakdown
Select any metric in the Metrics table to view its detailed cardinality information in a dedicated Cardinality Management page.
Bar charts
In the metric-specific view, you can check current cardinality usage with two bar charts:
- Metric cardinality budget (limit): Cardinality usage specific to the metric
- Account cardinality budget (limit): Overall cardinality usage for the account
Usage over time (for metrics)
The Usage over time chart displays the metric cardinality usage compared the limit of that metric. You can adjust the time picker to view different time periods of this individual metric's cardinality data.
팁
If a metric exceeds its cardinality limit, New Relic still populates information for the metric, but it may be slightly slower as it uses raw data instead of aggregated data.
Attributes table
The Attributes table has the list of top 20 attributes that majorly contribute to the metric's cardinality. To view the values of a specific attribute, click the options menu on the right side of the attribute row and select View attribute values.
Account pruning rules
A common method to maintain account cardinality within its limit is to implement pruning rules for non-critical attributes, which removes unnecessary data.
View account pruning rules
To view account pruning rules:
- Go to the Cardinality Management page of the account.
- From the Metrics table, click View Pruning Rules. A scrollable list of all pruning rules associated with the account appears.
팁
If no pruning rules exist in the account, the View Pruning Rules button doesn't appear.
Create a pruning rule
To create a pruning rule:
Go to the Cardinality Management page of the account.
From the Metrics table, select the metric in which you want to create a pruning rule. The Cardinality Management page of the metric opens.
Go to the Attributes table and select the attributes you want to prune.
Click Create pruning rule. A pop-up appears, providing details on how the pruning rule impacts the metric's cardinality.
Enter a rule name or description.
To create the pruning rule, click Submit.
중요
Pruning rules take effect immediately. Ensure you want to prune the data before creating a rule. You can delete a pruning rule after creating it, but the rule may prune some data before you remove it.
Delete a pruning rule
To delete a pruning rule:
- Go to the Cardinality Management page of the account.
- From the Metrics table, select View Pruning Rules.
- Scroll to the relevant pruning rule and delete it.
Per-metric limit adjustment
When all attributes of a specific metric are crucial and you approach your cardinality limit, you can adjust the per-metric limit to meet your needs. Typically, metrics have a default maximum cardinality of 100,000, but you can increase this up to 1 million per metric.
팁
Your account's overall cardinality limit remains unchanged. Therefore, setting all metric limits to 1 million without careful consideration could quickly exhaust your account's cardinality capacity.
Edit per-metric limit
- Go to the specific metric's Cardinality Management page.
- Select Adjust metric budget.
- Update the limit value to meet your needs. Make sure, it is less than or equal to 1 million.
The updated limit appears in the Cardinality Management page within a few minutes.
What's next
If you want to learn more about high cardinality, see Understand and query high cardinality metrics.