For Teachers and Admins
An Azure Cloud Lab can give each student their own Azure Databricks workspace. When the student starts the lab, Vocareum creates the workspace in the student's Azure resource group, and the student opens it from the lab panel.
This works differently from the other articles in this collection. They cover Vocareum's Databricks lab type. Databricks on Azure is set up as an Azure Cloud Lab instead: you pick Azure as the cloud vendor for the Part and Databricks as its lab service, and its settings are separate from the Databricks lab type's. If you haven't set up an Azure Cloud Lab before, start with Creating an Azure Cloud Lab.
Before you begin
Azure labs and Databricks must both be enabled for your organization. If Databricks Lab is not in the Lab Service list, contact Vocareum support.
Databricks works with Resource Group labs only. It is not offered for Subscription labs.
Set organization limits (Admins)
Org admins set the limits that every Databricks lab in the organization has to stay within. Instructors can configure their labs only inside these limits.
In the Control Center, open Azure, then Settings, and find the Databricks Lab Limits card:
Allowed Node Types: the VM sizes instructors can choose. Leave it empty to allow all of them.
Max Workers: the most worker nodes a cluster can have (default 8).
Max Clusters per User: how many clusters one student can create (default 3).
Max Idle Timer (minutes): the longest a cluster can sit idle before it shuts down (default 60, minimum 10).
Max DBUs per Hour: a cap on Databricks compute usage. Leave it blank for no cap.
Allowed Entitlements: which of Workspace access, Databricks SQL access and Unrestricted cluster create instructors can grant. The first two are allowed by default.
Allow Workspace Admin: whether instructors can make students admins of their own workspace. Off by default.
Click Save when you are done.
Set up a Databricks lab (Teachers)
Create the assignment and select Azure for the Part, as described in Creating an Azure Cloud Lab. Budgets and Interface options work the same as for any Azure lab.
Under Lab Service, choose Databricks Lab. The Databricks Workspace launch link is added for you.
Click Configure Lab Policy and set Worker Type, Driver Type, Max Workers, Idle Timer, DBUs per Hour, Max Clusters per User and Entitlements. Workspace Admin appears only if your admin allows it. Each field shows your organization's limit. New labs start with Standard_DS2_v2 workers and driver, 2 workers, a 30-minute idle timer, 1 cluster per user, and workspace and SQL access.
Click Done, then Save Part. The policy is not saved until you click Save Part.
If you copy the assignment to another course, its Databricks lab policy is copied with it.
NOTE: The Databricks lab policy is separate from Azure Lab Policies, which control which Azure resources students can create. You can use both on the same lab.
What students see
After the student clicks Start Lab, the Azure status indicator stays yellow (Initializing) while the workspace is set up. This can take a few minutes. When the workspace is ready, the indicator turns green and a Launch Databricks button appears. The Azure Details panel also shows an Open Workspace link. If a student tries to launch before the workspace is ready, they see a message asking them to wait.
Managing student labs (Teachers)
On the course's Teams page, a student's details show their Azure Cloud Lab status: Provisioning, Running, Not running or Failed. Click Terminate to end the student's session and delete their Azure lab account and its resources, including the workspace. The student can start a new lab afterwards.