Add example for a job writing to a Unity Catalog volume

knowledge_base/write_from_job_to_volume/.gitignore

@@ -0,0 +1 @@
+.databricks/

knowledge_base/write_from_job_to_volume/README.md

@@ -0,0 +1,20 @@
+# Save job result to volume
+
+This example demonstrates how to define and use a Unity Catalog Volume in a Databricks Asset Bundle.
+
+Specifically we'll define a `hello_world_job` job which writes "Hello, World!"
+to a file in a Unity Catalog Volume.
+
+The bundle also defines a Volume and the associated Schema in which the Job writes text to.
+
+## Prerequisites
+
+* Databricks CLI v0.236.0 or above
+
+## Usage
+
+Update the `host` field under `workspace` in `databricks.yml` to the Databricks workspace you wish to deploy to.
+
+Run `databricks bundle deploy` to deploy the job.
+
+Run `databricks bundle run hello_world_job` to run the job and store the results in UC volume.

knowledge_base/write_from_job_to_volume/databricks.yml

@@ -0,0 +1,12 @@
+bundle:
+  name: write_from_job_to_volume
+
+include:
+  - resources/*.yml
+
+workspace:
+  host: https://e2-dogfood.staging.cloud.databricks.com
+
+targets:
+  dev:
+    default: true

knowledge_base/write_from_job_to_volume/resources/hello_world.job.yml

@@ -0,0 +1,16 @@
+resources:
+  jobs:
+    hello_world_job:
+      name: hello_world_job
+
+      # No job cluster is configured. The job will run on serverless compute. 
+      # You can explicitly configure job compute here if your workspace does
+      # not have serverless compute enabled.
+      tasks:
+        - task_key: hello_world_job_task
+          notebook_task: 
+            notebook_path: ../src/hello.ipynb
+
+      parameters: 
+        - name: file_path
+          default: /Volumes/main/${resources.schemas.hello_world_schema.name}/${resources.volumes.my_volume.name}/hello_world.txt

knowledge_base/write_from_job_to_volume/resources/hello_world.schema.yml

@@ -0,0 +1,5 @@
+resources: 
+  schemas: 
+    hello_world_schema:
+      catalog_name: main
+      name: ${workspace.current_user.short_name}_hello_world

knowledge_base/write_from_job_to_volume/resources/my_volume.volume.yml

@@ -0,0 +1,9 @@
+resources: 
+  volumes: 
+    my_volume:
+      catalog_name: main
+      # We use the ${resources.schemas...} interpolation syntax to force the creation
+      # of the schema before the volume. Usage of the ${resources.schemas...} syntax 
+      # allows Databricks Asset Bundles to form a dependency graph between resources.
+      schema_name: ${resources.schemas.hello_world_schema.name}
+      name: my_volume

knowledge_base/write_from_job_to_volume/src/hello.ipynb

@@ -0,0 +1,21 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "file_path = dbutils.widgets.get(\"file_path\")\n",
+    "dbutils.fs.put(file_path, \"Hello World!\", overwrite=True)"
+   ]
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}