Microsoft Updated DP-750 Exam Questions and Answers by amalia

Data lineage in Unity Catalog is captured automatically — but only when jobs and notebooks run on clusters that are Unity Catalog-aware. Specifically, clusters must use ' Shared ' or ' Single User ' access mode. Clusters set to ' No Isolation Shared ' or legacy ' High Concurrency ' mode do not emit lineage events to the Unity Catalog lineage service.

No instrumentation, logging code, or external tools are required. The lineage service operates transparently, intercepting read and write operations at the Spark plan level and recording the table-to-table and column-to-column relationships.

To review captured lineage, open Catalog Explorer, navigate to the table, and select the Lineage tab. This shows the upstream sources that populate the table and the downstream consumers that read from it — all as an interactive graph, with no additional tooling needed. This built-in visibility is one of the core governance benefits Unity Catalog provides.

The correct answer is A — INSERT INTO.

INSERT INTO appends all rows from the source to the target. Existing rows in table2 are untouched — the historical data stays intact — and the current year ' s rows from table1 are added. That is exactly what the requirement asks: ' preserve any existing data and minimise processing effort. '

Option B (CREATE TABLE AS SELECT) would fail if table2 already exists, or if used with CREATE OR REPLACE it wipes table2 entirely before writing — historical data gone. Option C (INSERT OVERWRITE) replaces the entire table content with only the table1 rows, also destroying table2 ' s existing data. Option D (CREATE OR REPLACE TABLE AS SELECT) explicitly drops and recreates the table, eliminating all historical records.

INSERT INTO is deliberately the simplest possible command here — one line, no special options, no risk of accidental data loss.

The Databricks CLI contains the bundle command group for validating, deploying, and running Declarative Automation Bundles. CLI commands can be executed consistently from developer terminals or CI/CD pipelines, making deployments automated, repeatable, and suitable for multiple target environments. The Databricks SDK for Python can manage workspace APIs programmatically, but it is not the standard bundle deployment interface required here. Git folders provide source-control integration inside the workspace but do not deploy bundle-defined resources. The Jobs UI supports interactive creation and management of jobs, which introduces manual steps and does not provide the same infrastructure-as-code workflow. Using commands such as databricks bundle validate and databricks bundle deploy -t dev directly satisfies the automated development deployment requirement. Microsoft Learn

groupBy

pivot

The groupBy( " Region " ) operation establishes one aggregation group for each geographical region in the DataFrame. The subsequent pivot( " Product " ) operation converts every distinct Product value into a separate output column. Finally, agg(F.sum( " Sales " )) calculates the total sales for every product within each region. The resulting DataFrame contains one row per region and one column for each product, with the corresponding aggregated sales value. cogroup combines grouped datasets and is not used for creating a conventional pivot table. coalesce controls the number of DataFrame partitions or replaces null values, depending on its context. The sum function is already provided inside the aggregation expression and therefore is not required in either blank.