Usage History

In Wherobots Cloud, Usage History allows you to monitor the progress of workloads (notebooks, SQL sessions, and job runs), understand resource consumption, review the execution timeline, and verify configuration settings. The page was previously called Workload History. Use Usage History to see how Notebook Runtimes, Job Runs, and SQL Sessions consume resources across your Organization, spot usage trends and spikes, track quota usage, review activity by user or service principal, and guide resource allocation and cost decisions. You can review and filter any workload executed within your Wherobots Organization, and manage Job Runs.

Before you start

Before using this feature, ensure that you have the following required resources:

A Wherobots account.

View Notebook Runtime, Job Run, and SQL Session workloads

For insight into your workload trends and individual Job Runs, go to Usage History in the left sidebar of Wherobots Cloud.

Hover for details: Place your cursor over any point on the chart to display a tooltip with the exact quota usage breakdown at that moment. Use this to investigate any spikes or trends in resource consumption.

Usage History views

Usage History provides two main views to analyze your workloads: Chart and Table.

Chart view

In chart view, you can see an overview of all your workloads, including Notebook Runtimes, Job Runs, and SQL Sessions. The chart displays quota usage as a percentage of your organization’s total computing power over time. The chart displays Quota Usage as a percentage on the y-axis over the selected time range on the x-axis. Each area on the chart is color-coded by workload type, and the legend on the right side of the chart identifies each type:

(blue)
(purple)
(green)

Use the following filters to choose which workloads appear:

Filter	Description
Time Range	View workloads from the last 24 hours, 7 days, 30 days, 60 days, or 90 days. Defaults to the last 30 days.
Owner	Filter workloads by the Organization member or service principal that started them.
Region	Filter workloads by the AWS region where they ran (e.g., aws-us-west-2, aws-eu-west-1, aws-us-east-1, etc.).
Type	Filter workloads by type: Notebook Runtimes, Job Runs, SQL Sessions, or RasterFlow Runs.
Status	Filter workloads by status: Pending, Running, Failed, Completed, Cancelled, or Timed Out.

The usage chart is only available for the last 24 hours, 7 days, or 30 days. If you select a longer time range, Wherobots Cloud hides the chart and displays the table only.

Table view

The table view provides a detailed list of all workloads executed within your Wherobots Organization, along with key information about each workload.

The Table view includes the following elements:

Column	Description
Name	The name or ID of the workload. For Notebook Runtime and SQL Session workloads, click the workload name to copy its underlying workload ID to your clipboard. A tooltip displays the ID inline. This is useful when you need the ID for support or debugging, since these workload types do not have a separate detail page on Wherobots Cloud, like Job Runs do.
Owner	The entity that triggered the workload (Service Principal or User).
Duration	The total duration of the workload.
Time Completed	The date and time when the workload completed. Workloads that are still pending or running display a dash.
Status	The current status of the workload: Pending, Running, Completed, Cancelled, Failed, or Timed Out. Sorting this column groups workloads by status, with workloads that finished as expected (Completed, Cancelled) separated from those that did not (Failed, Timed Out).
Region	The AWS region where the workload executed.
Runtime	The runtime size used (e.g., Tiny, Small, Medium, Large).
Cost ($)	The estimated cost of the workload, rounded to two decimal places. Costs above zero but below one cent display as `< $0.01`.
SUs	Spatial Units consumed by the workload so far, rounded to two decimal places. Amounts above zero but below 0.01 display as `< 0.01`. This information can be delayed by up to 20 minutes.

If your account is configured to the User role, and you use a service principal-managed API key generated by an Admin in your Wherobots Organization, actions are attributed to the service principal in Audit Logs, with the Admin shown as responsible solely for the API key creation.

Filter workloads

To locate specific workloads, you can use multiple filters:

Navigate to Usage History.
Use the Filter by name or id field to search for specific workloads.
Apply filters for Time Range, Owner, Region, Type, or Status as needed.

Manage Job Runs

Scheduled script executions also appear as Job Run workloads. For their tick journal, skipped or missed windows, and reruns, see Monitor and recover task schedules. In the Usage History view, click on any Job Run workload row to view more information about that specific execution. The detail page includes two tabs: Details and Logs.

Details tab

The Details tab provides an overview of the Job Run’s configuration, runtime environment, and execution metadata so you can:

Review Spark version, runtime size, and environment details
Verify scheduling, trigger source, and timeout configuration
Understand when the Job Run started, ended, and how long it ran

The Details tab includes the following information about the Job Run execution:

Field	Description
Job Run Name	The name of the Job Run, which is either user-defined or auto-generated.
Job Run ID	Unique identifier for this specific Job Run.
Status	Indicates if the Job Run is initializing (`Pending`), in progress (`Running`), finished (`Completed`), unable to be executed (`Failed`), manually stopped (`Cancelled`), or stopped because it exceeded its configured timeout (`Timed Out`).
Start Time	Indicates when the Job Run was initiated.
End Time	Indicates when the Job Run terminated.
Duration	Indicates how much time elapsed between the Start Time and End Time.
Triggered By	The entity that triggered the Job Run (e.g., Airflow, user email).
Runtime	The runtime size used for this Job Run (e.g., Tiny, Small, Medium, Large).
Runtime Version	The version of the runtime environment used for this Job Run. When you specify a version alias such as `latest` or `preview`, the resolved concrete version is shown in parentheses (for example, `latest (1.5.1)`).
Timeout	The maximum duration the Job Run is allowed to run before being automatically stopped, in seconds. The default is `3600`.
Region	The AWS region where the Job Run executed. Different regions may have different performance characteristics and costs.
Python File	The URI of the Python file executed in this Job Run.
Environment	The Spark configuration details for this Job Run.
Cancel Button	Allows you to stop an ongoing Job Run.

Consumption metrics

The Details tab also shows the consumption metrics for the Job Run:

Metric	Description
Cost ($)	The estimated cost of the workload execution, rounded to two decimal places. Costs above zero but below one cent display as `< $0.01`.
Spatial Units Consumed	Indicates the accumulated Spatial Unit Consumption for this workload, rounded to two decimal places. Amounts above zero but below 0.01 display as `< 0.01`.
Max CPU Utilization	The highest CPU utilization observed during the workload, as a percentage.
Max Memory Utilization	The highest memory utilization observed during the workload, as a percentage.

Logs tab

The Logs tab provides real-time access to execution logs for your Job Run including information about:

File downloads and uploads
Spark submit commands
Spark version and environment details
Warnings and error messages
Application execution progress

The Logs tab shows the Job Run name, ID, and status, as on the Details tab, and the following:

Element	Description
Spark UI	A link to the Spark UI for this Job Run. Available while the Job Run is in the `Running` state.
Spark event logs	A folder icon button next to the Spark UI link opens the Spark event logs for this Job Run on the Cloud Storage page. Use these logs with tools like the Spark History Server to analyze a Job Run’s performance after it has finished. Available when the Job Run produced event logs in your Wherobots-managed storage.
Search Logs	Search through log entries using `Cmd`/`Ctrl` + `F`.
Log Timeline	View timestamped log entries showing the execution progress.
Copy Logs	Copy all log entries to clipboard.
Download Logs	Download the complete log file for offline analysis.

Compare workload performance

To compare workload performance with different runtime types:

Run a workload. For more information on initializing a job and runtime selection, see WherobotsRunOperator.
Run that same workload again with either a larger or smaller runtime type.
Review the Detail pages for each executed Job Run to compare performance metrics, specifically:
- Duration
- Max CPU Utilization
- Max Memory Utilization
- Cost($)
- SUs (Spatial Units)

Cancel a workload

Wherobots cancels workloads based on your specification of timeoutseconds in the Job Runs REST API Schema, which defaults to 3600, but you can also cancel a workload in Wherobots Cloud. To cancel a workload within Wherobots Cloud, do the following:

In Usage History, locate the Job Run you wish to cancel.
Click on the Job Run to open its detail page.
Click the Cancel button.

You can only cancel workloads that are in the Pending or Running state.

Limitations

The following limitations apply to Usage History:

90-day data retention: Usage History data, including Spatial Unit Consumption, Max CPU Utilization, and Max Memory Utilization metrics, is only kept for 90 days.
Slightly delayed cost and SU data: Spatial Unit Consumption and Cost can take several minutes to display on a Job Run’s Detail page.
Quota usage is percentage-based: The chart displays quota usage as a percentage of your organization’s total computing power.

Start here

Notebooks

Editors and AI assistants

VS Code extension

MCP server and agent skills

dbt

Apache Airflow

Storage management

Task schedules

Deployment and monitoring

Before you start

View Notebook Runtime, Job Run, and SQL Session workloads

Usage History views

Chart view

Table view

Filter workloads

Manage Job Runs

Details tab

Consumption metrics

Logs tab

Compare workload performance

Cancel a workload

Limitations