The Cluster Overview page displays key metrics about your CockroachDB serverless cluster.
The Cluster settings panel displays the settings you chose during cluster creation.
|Cloud||The cluster's cloud provider.|
|Plan type||The plan type used to create the cluster.|
|Region||The cluster's region.|
Usage this month
The Usage this month panel displays your cluster usage statistics for the current month. You can click Estimate usage cost to open a tool that estimates your monthly costs based on your workload during a selected time frame.
|Spend limit||The spend limit you entered when creating the cluster. Click the edit icon to change the spend limit.|
|Storage||The amount of data currently stored in the cluster. This value does not account for compression or replication. The total available storage is determined by your spend limit and Burst RU usage. For details, see CockroachDB Cloud Architecture.|
|Request Units||The CPU and I/O resources being used by queries on the cluster, and whether throttling is in effect. The total available RUs are determined by your spend limit and storage usage. For more context, see CockroachDB Cloud Architecture.|
Cluster statistics panel
This panel displays operational statistics for your cluster.
|Throughput||The cluster throughput in queries per second (QPS).|
|P99 latency||The 99th percentile of service latency. This value indicates the time within which the cluster executed 99 percent of queries over the last minute.|
|Open sessions||The number of open sessions in the cluster.|
Cluster Overview metrics
These time-series graphs display important cluster metrics over time.
You can use the timeframe selector above the graphs to view data for a specific time period.
The Storage Used graph displays the amount of data being stored in the cluster over time. This is the logical number of live bytes and does not account for compression or replication.
The Request Units graph displays the CPU and I/O resources being used by queries on the cluster. This is measured in Request Units per second at each timestamp. Simple queries consume few RUs, while complicated queries with many reads and writes consume more RUs.