View billable usage using the account console

This article describes how to use the Usage page in the account console to view usage data across workspaces in your account. You can also download billable usage logs using the Account API.

You can also access and query billable usage data using Databricks system tables (Public Preview).

You can view usage in the account console by either of these units:

  • Databricks Unit (DBU): A unit of processing capability per hour, billed on per-second usage.

  • Estimated cost (in $USD currency): Cost in dollars is estimated on a linear cost estimate per DBU based on standard cost-per-DBU pricing for your tier for each type of usage. Actual cost varies depending on your exact contract. Adjust cost-per-DBU for each SKU to change the cost estimate using the settings panel.

Access the Usage page

Go to the account console and click the Usage icon.

Usage graph

At the top of the page, a graph shows your account’s usage in DBUs or the estimated cost in $USD. Use the $USD/DBU picker to toggle between these views.

You can also use the aggregation picker to browse the data by:

  • Total usage: A single line for the account’s total usage across all workspaces and SKUs. This is the default selection.

  • Workspace: A line for each workspace’s usage.

  • SKUs: A line for each SKU’s usage. The set of SKUs shown for your account depends on your contract. SKUs also depend on how you use Databricks. See Pricing.

  • Tags: A line for each resource that includes the selected tag.

If there are more than 10 workspaces, SKUs or tag values, the chart displays the nine with the highest usage. The usage of the remaining workspaces, SKUs or tag values is aggregated and displayed in a single line. This single line is marked as “Combined remaining …(s)”.

If you browse data by tags, usage without the selected tag is aggregated and displayed in a single line labeled “(… tag absent).”

Filter by workspace, tag, or SKU

You can use the graph’s legend or the line picker to filter out the graph lines:

  • To show / hide a line, click on its radio button in the graph’s legend or its checkbox in the line picker.

  • To only show a single line, double-click on its radio button in the graph’s legend or its checkbox in the line picker. You can then double-click it again to restore the view and show all lines.

The actual cost varies depending on your contract. You can adjust the cost-per-DBU for each SKU in the settings panel to change the cost estimate.

Usage details

In the Usage details panel, you can view a list of detailed usage in DBU or the estimated cost in $USD in a table format. This panel’s data is always aggregated by workspace. The cost is also broken down into SKU groups. To view usage as totals (aggregated) or by SKU, use the graph in the settings panel instead.

To toggle between estimated cost and DBUs, use the $USD/DBU picker.

Use the date range picker to specify an exact time range or choose among pre-defined options such as Last 6 months. The time ranges are in UTC time.

If you have many workspaces, type the workspace name into the search field to filter results.

The actual cost varies depending on your contract. You can adjust the cost-per-DBU for each SKU in the settings panel to change the cost estimate.

Pricing

You can set your actual pricing per DBU by clicking on the vertical ellipsis in the top right corner of the Usage page to open the settings panel. Use this panel to adjust estimated pricing per DBU for both the usage and the usage details table on the Usage page.

Each SKU is listed separately. The set of SKUs shown for your account depends on your contract. How usage applies to SKU depends on what you do with Databricks. For example, it depends whether a notebook is run using an automated cluster or as a scheduled job.

Usage downloads

There are two types of usage downloads available. You can download the aggregated usage data displayed in your current graph or download overall usage details by date range.

Download aggregated usage data

To download aggregated usage data, click the download button in the top right corner of the graph. This downloads the data currently aggregated on the page. The CSV file will include the 999 workspaces, SKUs, or tag values with the highest usage.

Usage data for unselected workspaces, SKUs, and tags is aggregated into groups labeled “Combined remaining …(s)” and “(… tag absent).”

Download unaggregated usage data

In the settings panel you can download all usage data by date. Open the panel by clicking on the kebab menu in the top right corner of the Usage page. Scroll to the bottom of the panel.

  1. Select a time range from the month range picker.

  2. To include usernames, enable Include usernames (email).

  3. Click Download.

Analyze usage logs

For information about how to analyze these files using Databricks, see Billable usage log schema.

Download usage logs with the Account API

You can download billable usage logs using the billable usage log download API.

Configure automated usage log delivery

To configure automated usage log delivery to an S3 bucket, see Deliver and access billable usage logs.