Data Analyst Associate · Domain 1: Understanding of Databricks Data Intelligence Platform (Lakehouse Platform) · 11% of the exam
Task 1.1: Describe the core components of the Databricks Data Intelligence Platform (Lakehouse Platform), including Mosaic AI, Lakeflow Declarative Pipelines (DeltaLive tables), Lakeflow Jobs, Data Intelligence Engine, Delta Lake, Lakebase, Unity Catalog, and Databricks SQL; Describe the role and features of Databricks Marketplace
What Delta Lake, Unity Catalog, Databricks SQL, Lakeflow pipelines and Jobs, Lakebase and the AI features each do, and how Marketplace listings, private exchanges and OpenSharing bring data products into a workspace.
Study it
The platform's components: Delta Lake, Unity Catalog, Databricks SQL, Lakeflow and Lakebase
Lesson coming
Databricks Marketplace: listings, private exchanges and shared catalogs
Lesson coming
Sample questions
Try each one before opening the answer. Every option is explained, with the Databricks documentation page that proves it.
Question 1 · choose 1
A team built several pipelines with Delta Live Tables two years ago. Today the documentation calls the product Lakeflow pipelines, built on Apache Spark Declarative Pipelines (the exam guide says Lakeflow Declarative Pipelines). The team's SQL and Python pipeline code has not changed. What must the team do so that the existing pipelines keep running?
ANothing is required; the existing code still works as Lakeflow pipelines
BRewrite each pipeline as a Lakeflow job with one notebook task per table
CRecreate each table as a standalone materialized view in Databricks SQL
DExport the code and run it on open-source Apache Spark Declarative Pipelines
Show the answer and why
ANothing is required; the existing code still works as Lakeflow pipelines
Correct
Databricks states that no migration is required to use Lakeflow pipelines when you have used DLT before: the code still works.
BRewrite each pipeline as a Lakeflow job with one notebook task per table
Incorrect
Lakeflow Jobs orchestrates tasks; the pipelines themselves need no rewrite to keep running.
CRecreate each table as a standalone materialized view in Databricks SQL
Incorrect
Nothing in the rename asks for the pipelines' tables to be recreated elsewhere; the existing pipelines keep working.
DExport the code and run it on open-source Apache Spark Declarative Pipelines
Incorrect
Lakeflow pipelines run on Databricks and are interoperable with the open framework; moving off Databricks is not needed to keep them running.
The product formerly known as Delta Live Tables (DLT) has been updated to Lakeflow pipelines, and Databricks says there is no migration required: existing code still works. New names (for example, from pyspark import pipelines as dp) are recommended for new work but optional.
An analyst runs CREATE TABLE main.sales.stores AS SELECT * FROM read_files('/Volumes/main/landing/files/stores.csv', format => 'csv') and does not add a USING clause. In which format does Databricks store the new table?
AAs Apache Iceberg tables, the format that Unity Catalog manages
BAs a Delta Lake table
CAs CSV files, the format of the file the table was read from
DAs plain Parquet files with no transaction log
Show the answer and why
AAs Apache Iceberg tables, the format that Unity Catalog manages
Incorrect
Without a format clause, tables on Databricks are Delta Lake tables; Iceberg is not the default.
BAs a Delta Lake table
Correct
Delta Lake is the default format for all operations on Databricks, so a table created without a format clause is a Delta table.
CAs CSV files, the format of the file the table was read from
Incorrect
The source file's format does not carry over to the new table; the default table format applies.
DAs plain Parquet files with no transaction log
Incorrect
Delta Lake extends Parquet data files with a transaction log; the new table is a Delta table, not plain Parquet.
"Delta Lake is the default format for all operations on Databricks. Unless otherwise specified, all tables on Databricks are Delta Lake tables." Delta Lake is open source software that extends Parquet data files with a file-based transaction log for ACID transactions.
A retailer's analysts work in Microsoft Power BI. The company also has one Databricks workspace, which is not enabled for Unity Catalog. They find two free Databricks Marketplace listings they want: a Data set listing of weather history and an AI model listing. Which statement describes what they can do?
AGet both listings in Power BI once the provider approves the request
BGet neither listing until they enable Unity Catalog on their workspace
CGet both listings in Power BI without using any Databricks workspace, for the request as well
DGet the weather data in Power BI; the AI model listing needs a workspace enabled for Unity Catalog
Show the answer and why
AGet both listings in Power BI once the provider approves the request
Incorrect
Through external platforms and workspaces without Unity Catalog, only the Data set product type can be shared; approval does not change that.
BGet neither listing until they enable Unity Catalog on their workspace
Incorrect
Data set listings can be accessed from external platforms such as Power BI through Databricks-to-Open sharing connectors, without Unity Catalog.
CGet both listings in Power BI without using any Databricks workspace, for the request as well
Incorrect
A request for a data product needs a Databricks workspace, and the AI model listing is not available on external platforms.
DGet the weather data in Power BI; the AI model listing needs a workspace enabled for Unity Catalog
Correct
Only Data set (tabular data) listings can be accessed with external platforms or workspaces without Unity Catalog; other product types need a Unity Catalog-enabled workspace.
Marketplace consumers without a Unity Catalog-enabled workspace can use Databricks-to-Open sharing connectors for platforms such as Power BI, Excel, pandas and Apache Spark, but only for the Data set product type. Requesting a data product always goes through a Databricks workspace; reading the shared data afterwards does not need one.
Which component of the Databricks Data + AI Platform (the exam guide: Data Intelligence Platform) is a fully managed Postgres database for online transactional processing (OLTP), integrated with the platform?
ADelta Lake
BDatabricks SQL
CLakeflow Jobs
DLakebase
Show the answer and why
ADelta Lake
Incorrect
Delta Lake is the optimized storage layer for tables in the lakehouse, built on Parquet files and a transaction log.
BDatabricks SQL
Incorrect
Databricks SQL is the data warehousing offering that runs SQL on SQL warehouses.
CLakeflow Jobs
Incorrect
Lakeflow Jobs schedules and orchestrates data processing tasks.
DLakebase
Correct
Lakebase is an OLTP database fully integrated with the platform: a fully managed Postgres database.
Lakebase is an online transactional processing (OLTP) database that is fully integrated with the Databricks Data + AI Platform; the fully managed Postgres database lets you create and manage OLTP databases stored in Databricks-managed storage.
Which component is described as the unified governance layer for data and AI in Databricks, enforcing access control when a table is queried?
AUnity Catalog
BLakeflow pipelines
CLakebase
DPhoton
Show the answer and why
AUnity Catalog
Correct
Unity Catalog is the unified governance layer for data and AI built into Databricks and enforces access control on queries.
BLakeflow pipelines
Incorrect
Lakeflow pipelines build batch and streaming data pipelines; they are not the governance layer.
CLakebase
Incorrect
Lakebase is a managed Postgres OLTP database.
DPhoton
Incorrect
Photon is the vectorized query engine that speeds up workloads.
Unity Catalog is the unified governance layer for data and AI built into Databricks. When enabled for a workspace it operates beneath every data and AI interaction, enforcing access control when you query a table or call a model.
A data company wants to list free sample datasets on Databricks Marketplace. Its Databricks account has several workspaces, and none of them is enabled for Unity Catalog yet. Which setup meets the Marketplace provider requirements for the account and its workspaces?
AUnity Catalog enabled on every workspace in the account first
BAny workspace on the Standard plan, as free listings skip OpenSharing
CA Premium account and one Unity Catalog workspace, even one just for listings
DNo workspace, as providers create listings on the Open Marketplace site
Show the answer and why
AUnity Catalog enabled on every workspace in the account first
Incorrect
The provider requirements say you do not need to enable all of your workspaces for Unity Catalog.
BAny workspace on the Standard plan, as free listings skip OpenSharing
Incorrect
Marketplace uses OpenSharing, which needs a Unity Catalog-enabled workspace, and the account must be on the Premium plan or above.
CA Premium account and one Unity Catalog workspace, even one just for listings
Correct
Providers need an account on the Premium plan or above and a workspace enabled for Unity Catalog; one created only to manage listings is enough.
DNo workspace, as providers create listings on the Open Marketplace site
Incorrect
Listings are managed from the provider console, which opens in a Unity Catalog-enabled workspace.
Databricks Marketplace uses OpenSharing, which requires a workspace enabled for Unity Catalog. A provider therefore needs a Databricks account on the Premium plan or above and a Unity Catalog-enabled workspace; not every workspace has to be enabled, and one can be created just to manage Marketplace listings.
A team installed a Marketplace data product, and several dashboards query its shared catalog. The team now removes the installed data product. What happens?
AThe shared catalog is deleted; dependent dashboards need updating
BThe shared catalog stays, but its tables stop receiving updates
CThe catalog's data is copied into a managed catalog the team owns
DNothing changes until the provider deletes the listing
Show the answer and why
AThe shared catalog is deleted; dependent dashboards need updating
Correct
Removing an installed data product deletes the shared catalog, and Databricks reminds you to update or delete dependent queries, dashboards and notebooks.
BThe shared catalog stays, but its tables stop receiving updates
Incorrect
The shared catalog does not stay behind; removing the product deletes it.
CThe catalog's data is copied into a managed catalog the team owns
Incorrect
Removal does not copy data; the shared catalog is deleted.
DNothing changes until the provider deletes the listing
Incorrect
The removal takes effect in the team's workspace; the shared catalog is deleted.
When you remove an installed Marketplace data product, the shared catalog is deleted from your workspace, so queries, dashboards and notebooks that depend on it must be updated or deleted.