What you will be able to do
- Explain what happens during each scheduled dashboard update
- Create a dashboard schedule with frequency, start time, time zone and the optional advanced settings
- Predict when a schedule with a defined start time will run
- Edit, pause, resume, delete or immediately run a schedule, and read its run history
Key concept
Scheduled dashboard update — A schedule is a recurring cadence attached to a dashboard. On each run it re-executes the SQL behind every dataset and stores the results in the cache, so viewers open the dashboard to fresh, precomputed data instead of waiting for queries to run.
1.What a scheduled run actually does
An AI/BI dashboard is built on datasets, and each dataset is defined by SQL. If nothing refreshes them, the numbers a viewer sees depend on when someone last ran those queries. A schedule fixes that. You give the dashboard a recurring cadence, and on every run Databricks executes all the SQL that defines its datasets, not only the datasets behind the page someone happens to be looking at.
The results are not discarded after the run. They populate the query result cache, and that cache is what makes the dashboard open quickly. In the documentation's words, a schedule "pulls the freshest data and, for dashboards published with shared data permissions, refreshes the shared cache." A schedule therefore does two jobs: it keeps the data fresh, and it moves query time away from the moment a viewer opens the dashboard. When you think about scheduling, think about both. Freshness is the reason most people set one up, and the faster first load is the benefit they tend to forget.
Checkpoint 1 of 6· Check yourself
A dashboard has five datasets spread across three pages. What runs during each scheduled update?
Each scheduled update runs the SQL for every dataset on the interval you set, and the results go into the result cache.
“All SQL logic that defines datasets runs on the designated time interval.”Source: docs.databricks.com
2.Creating a schedule and its advanced settings
You create schedules from the dashboard itself. Click Schedule in the upper-right corner of the dashboard. Use the drop-down selectors to set the frequency, start time and time zone. If the drop-downs can't express the cadence you need, select the Show cron syntax checkbox and edit the schedule in Quartz Cron Syntax. Expand Advanced settings for the optional choices, then click Create. The Schedules dialog then lists your new schedule along with any others that already exist on the dashboard. A dashboard can have more than one schedule.
Checkpoint 2 of 6· Put it in order
Put the steps for creating a dashboard schedule in order
- 1.Optionally open Advanced settings to set a name, warehouse, filters, subscribers or attachments
- 2.Click Create and review the schedule in the Schedules dialog
- 3.Click Schedule in the upper-right corner of the dashboard
- 4.Use the drop-down selectors to set frequency, start time and time zone
Open Schedule first, set the timing, optionally adjust Advanced settings, and finish with Create. The Schedules dialog then shows every schedule on the dashboard.
“Use the drop-down selectors to specify the frequency, start time, and time zone.”Source: docs.databricks.com
Exam scenarios often turn on the advanced settings, so learn each one. The table below lists them.
| Setting | What it controls |
|---|---|
| Name | A name identifier for the schedule |
| SQL warehouse | A different warehouse for scheduled runs. By default, the warehouse used to create and run the dashboard is used |
| Use current filter selections | Applies the dashboard's current filter selections during the scheduled run |
| Custom email subject | The subject line of subscription email notifications |
| Subscribers | Users or notification destinations notified after each scheduled run |
| Attachments | Include pages (PDF, in selected order) and Include data (CSV, TSV, or Excel per widget) |
Two of these settings change which data the run produces, not only how it is delivered. The warehouse setting lets you send scheduled work to a different SQL warehouse from the one you use while authoring. Use current filter selections means the run uses the filters as they are currently set, rather than ignoring them.
Checkpoint 3 of 6· Check yourself
An analyst wants a dashboard's scheduled refresh to run on a separate warehouse instead of the one used for interactive work. What should they do?
The dashboard's own warehouse is only the default. Advanced settings lets you pick a different SQL warehouse for scheduled runs.
“SQL warehouse: Choose a different SQL warehouse for scheduled runs.”Source: docs.databricks.com
Checkpoint 4 of 6· Exam question
Which control does a data analyst use to begin configuring an automatic refresh schedule for an AI/BI dashboard in Databricks?
Correct answer: A — The Schedule button in the dashboard's upper-right corner, which opens the Add schedule dialog.
- A. This is correct because opening a dashboard exposes a dedicated Schedule control that launches the dialog for adding a new refresh schedule. From there the analyst sets frequency, time, time zone, and warehouse before saving.
- B. This is incorrect because that dialog governs who can view or edit the dashboard, not when its data refreshes. Permissions and refresh scheduling are separate configuration surfaces.
- C. This is incorrect because that panel only shows past query executions and their durations. It has no controls for creating or editing a recurring refresh schedule.
- D. This is incorrect because that page manages when compute automatically starts, stops, or scales, not when a dashboard's datasets are refreshed. Alerts on a warehouse also monitor query conditions, not dashboard content freshness.
3.How the start time anchors the cadence
The start time is not just when the first run happens. It anchors the whole day. The documentation's example is a schedule set to every 4 hours starting at 15:10. It runs at 15:10, 19:10, 23:10 and so on, continuing until 15:09 the following day. At that point the cadence resets and begins again at 15:10. So you should read a schedule as "every N hours, counted from the start time, restarting at the start time each day", not as a counter that drifts forward forever. The time zone setting matters for the same reason: it decides which 15:10 the schedule means.
Five hours doesn't divide evenly into 24. The runs would be 08:00, 13:00, 18:00, 23:00 and 04:00. Instead of continuing to 09:00 the next day, the cadence resets and runs again at 08:00. The day restarts at the anchor rather than drifting.
Checkpoint 5 of 6· Check yourself
A schedule runs every 4 hours starting at 15:10. Which statement about it is correct?
The cadence runs through the day from the start time and resets at the start time the next day.
“The cadence then resets and begins again at 15:10.”Source: docs.databricks.com
Sources1
4.Editing, pausing, deleting, running now and checking history
To manage an existing schedule, click Schedule again to list the dashboard's schedules, then open the kebab menu to the right of the one you want. From there you can choose:
- Edit reopens the scheduling UI so you can adjust the schedule and save it. - Pause / Resume turns the schedule off or on. A paused schedule keeps its configuration and subscriber settings, but the dashboard neither updates nor sends notifications until you resume it. - Delete removes the schedule and all of its subscriber settings. - Run now triggers a run immediately.
Running a schedule now updates the dashboard and sends notifications without changing its regular cadence, so the 12:00 run still happens. You don't need to wait for the next notification to see whether a schedule is working. The schedule list shows run history indicators to the right of each schedule's details, with the most recent run on the right. Hover over an indicator to see the run ID, start time, end time and status. Each indicator shows one of these outcomes: succeeded, failed, or skipped because the schedule had already reached its limit for concurrent runs. Runs that are in progress or waiting to start also appear, and a dash marks a position where no run happened.
Checkpoint 6 of 6· Match them up
Match each schedule action to its effect
Tap a term, then the definition that fits it.
Pausing is reversible and keeps the subscriber list. Deleting removes both the schedule and its subscribers. Run now is a one-off run that leaves the cadence unchanged.
“Paused schedules retain their configuration and subscriber settings”Source: docs.databricks.com
Exam traps
Each one states something that sounds right. Open it to see what is actually true.
1.Pausing a schedule over a holiday throws away its subscriber list, so you'll have to rebuild it afterwards.Why is that wrong?
Pausing keeps the configuration and subscribers. Only Delete removes the subscriber settings.
Covered in Editing, pausing, deleting, running now and checking history
2.Scheduled runs always use the SQL warehouse the dashboard was built with.Why is that wrong?
That warehouse is only the default. Advanced settings lets you choose a different SQL warehouse for scheduled runs.
Sources
Every claim above is drawn from one of these pages, quoted as it was written on the date shown.
- 1.
“Results populate the query result cache and help to improve initial dashboard load time.”
↩︎ What a scheduled run actually does“pulls the freshest data and, for dashboards published with shared data permissions, refreshes the shared cache.”
↩︎ What a scheduled run actually does“Click Schedule in the upper-right corner of the dashboard.”
↩︎ Creating a schedule and its advanced settings“Optionally, select the Show cron syntax checkbox to edit the schedule in Quartz Cron Syntax.”
↩︎ Creating a schedule and its advanced settings“Select Use current filter selections to apply the current filter selections during a scheduled run.”
↩︎ Creating a schedule and its advanced settings“For schedules with a defined start time, the refresh cadence begins at that time each day.”
↩︎ How the start time anchors the cadence“Running a schedule now updates the dashboard and sends notifications without changing its regular cadence.”
↩︎ Editing, pausing, deleting, running now and checking history“skipped because the schedule had already reached its limit for concurrent runs”
↩︎ Editing, pausing, deleting, running now and checking history“All SQL logic that defines datasets runs on the designated time interval.”
↩︎ Key concept“Delete: Removes the schedule and all subscriber settings.”
↩︎ Exam trap 1“By default, the SQL warehouse used to create and run the dashboard is also used in the scheduled run.”
↩︎ Exam trap 2“All SQL logic that defines datasets runs on the designated time interval.”
↩︎ Checkpoint“Use the drop-down selectors to specify the frequency, start time, and time zone.”
↩︎ Checkpoint“SQL warehouse: Choose a different SQL warehouse for scheduled runs.”
↩︎ Checkpoint“The cadence then resets and begins again at 15:10.”
↩︎ Checkpoint“Paused schedules retain their configuration and subscriber settings”
↩︎ Checkpoint - 2.
“If a schedule is set, all datasets refresh according to the schedule, and those results are cached.”
↩︎ What a scheduled run actually does“To reliably refresh the dashboard cache, configure a dashboard schedule.”
↩︎ What a scheduled run actually does - 3.
“You can schedule the dashboard to refresh at an interval automatically.”
↩︎ Creating a schedule and its advanced settings - 4.
“Schedules and subscriptions: Set up automated refresh schedules and email or Slack notifications for stakeholders”
↩︎ Editing, pausing, deleting, running now and checking history