Microsoft DP-750 Implementing Data Engineering Solutions Using Azure Databricks Exam Practice Test

The right ingestion tool depends on the source characteristics:

File-based telemetry and maintenance data → Auto Loader (cloudFiles). It monitors ADLS Gen2 for new arrivals, handles schema inference and evolution for the frequent schema drift, and requires minimal operational effort.

Real-time telemetry from Event Hubs → Spark Structured Streaming with the azure-eventhubs-spark connector. This provides exactly-once semantics and checkpoint-based recovery, satisfying 'resume processing after failures without reprocessing.'

Structured maintenance data from PostgreSQL → JDBC connector. Databricks supports direct JDBC reads from relational databases with pushdown predicates.

Daily CSV ERP extracts (50–100 GB) → COPY INTO or Auto Loader. Both support idempotent incremental batch loading into Delta tables with minimal code.

The correct answer is A. Photon is Azure Databricks' native vectorized query engine, written in C++, designed to accelerate data ingestion and SQL-heavy workloads significantly over the standard Spark JVM path. Enabling it on a job compute cluster directly addresses Contoso's requirement for 'fast and consistent performance for BI workloads' and 'production ingestion workloads that can scale automatically during telemetry spikes.'

Photon integrates transparently — no code changes are needed — and pairs well with autoscaling job clusters to handle the bursty 40,000-sensor telemetry load.

Option B contradicts the isolation requirement: Contoso explicitly needs production and development separated, not merged onto shared compute. Option C with a fixed large node gives peak capacity at all times, driving up costs even during quiet periods. Option D disabling autoscaling is the opposite of what's needed — telemetry spikes require elastic scaling, not a locked node count.

The solution requires spark.readStream with format('cloudFiles') for Auto Loader, paired with .writeStream using mergeSchema=true and a checkpointLocation.

Auto Loader's cloudFiles source incrementally processes new JSON files without rescanning the entire directory. The mergeSchema option handles schema drift — when sensors add new fields, the target Delta table schema expands automatically instead of throwing a parse error. This directly addresses Contoso's requirement to 'support schema drift.'

The checkpointLocation is what gives the pipeline its resilience. Databricks writes the stream's committed offset and schema state to that path. If the cluster restarts, the engine reads the checkpoint and picks up exactly where it left off — no events are reprocessed, satisfying 'exactly-once semantics' and 'resume processing after failures without reprocessing the data.'

Without a checkpoint, the stream would restart from the beginning on every cluster bounce, which is precisely the problem Contoso is trying to eliminate.

The correct answer is D. Breaking the pipeline into separate tasks for ingestion, cleansing, and curation is the foundation of well-designed Lakeflow Jobs pipelines. Each task should own one responsibility — when a task does too much, debugging a failure becomes a hunt through unrelated code, and retry logic becomes expensive because you re-execute work that already succeeded.

Contoso's planned changes explicitly call for 'a clear execution order and dependencies' and 'orchestrate multi-step ingestion and transformation workflows.' Separate tasks map directly to those goals: Lakeflow Jobs tracks each task's status independently, so if cleansing fails, ingestion doesn't re-run.

Option A bundles everything into one notebook, which means a curation bug forces a full re-ingestion. Option B copies logic three times — any future change must be applied in triplicate, which is a maintenance hazard. Option C forces everything through SQL MERGE, which is the wrong tool for raw-event ingestion and doesn't address cleansing or schema drift.

The correct mapping is SCD Type 1 for equipment metadata and SCD Type 2 for IoT sensor ownership history.

SCD Type 1 overwrites the existing record whenever an attribute changes — no history is kept. Contoso's requirement for equipment metadata (name, manufacturer, model, commissioning date) states 'historical values are NOT required,' which is the textbook definition of Type 1. A MERGE INTO with WHEN MATCHED THEN UPDATE handles this cleanly in Delta Lake.

SCD Type 2 creates a new row for each change, preserving the full history through effective-date or version columns. Contoso requires that 'analysts must track the full history of ownership' as sensors change hands over time — that full audit trail is only possible with Type 2. Type 3 (keeping just the previous value in an extra column) would lose earlier ownership records, so it doesn't satisfy the 'full history' requirement.

The correct answer is B — No.

df.fillna(0, subset=['order_amount']) does not remove null rows. It replaces null values in order_amount with the integer 0, leaving those rows in the DataFrame with 0 as their order_amount value. The row count is unchanged — nulls are imputed, not dropped.

The requirement is to exclude rows where order_amount is null, meaning those rows should not appear in the result at all. fillna is a data imputation function — it fills gaps with default values, which is a different operation entirely from filtering. After fillna, a downstream process would see order_amount = 0 and might treat it as a valid zero-value order rather than recognising it was originally null.

The correct approach is df.dropna(subset=['order_amount']) or df.filter(df.order_amount.isNotNull()), both of which physically remove null rows from the resulting DataFrame.

The correct answer is C — the Databricks CLI.

Databricks Asset Bundles are deployed using the Databricks CLI (v0.205+) with the commands databricks bundle validate, databricks bundle deploy, and databricks bundle run. The CLI reads the databricks.yml file, resolves the target environment settings, and creates or updates all declared resources (jobs, pipelines, apps) in the workspace. This is the documented, supported deployment mechanism for DABs and integrates naturally into CI/CD pipelines.

Option A (Azure Developer CLI / azd) is a general Azure IaC tool with no native understanding of Databricks bundles. Option B (Git folders) syncs notebook and file content from a Git repository into a workspace — useful for code, but it doesn't deploy Lakeflow Jobs, SDP pipelines, or bundle configurations. Option D (Azure CLI) manages Azure infrastructure (resource groups, storage accounts, etc.) and has no native DAB deployment support.

The correct answer is B — No.

Lakeflow Connect is an ingestion service that physically copies data from external databases into Delta tables managed by Databricks. It's designed for scenarios where you want a replicated, writable Delta copy of external data — essentially a CDC-based ingestion pipeline.

That's the opposite of what's required here. The requirement states 'the data is NOT copied into Databricks-managed storage.' Lakeflow Connect would create Delta tables in Databricks and copy DB1's data into them — a direct violation.

Additionally, Lakeflow Connect creates Databricks-native Delta tables rather than exposing DB1's original schemas and tables as virtual objects. Analysts querying through a Lakeflow Connect pipeline are querying a replicated copy, not the live source.

For zero-copy, live query federation of an external SQL Server into Unity Catalog, Lakehouse Federation (foreign catalog) is the correct tool.

The correct answer is B — No.

A native catalog in Unity Catalog is a standard Databricks-managed catalog. Creating one provisions a metadata namespace inside Unity Catalog, but it has no connection whatsoever to DB1 or any external SQL Server database. There is no mechanism to point a native catalog at an external database — it simply doesn't serve that purpose.

The requirement is to expose an external SQL Server's schemas and tables inside Unity Catalog without copying the data. That requires a foreign catalog, which is created through Lakehouse Federation using a registered connection to the external database. A foreign catalog acts as a read-only, virtual mirror of the external database — queries run against the live external data in place.

Creating a native catalog and then trying to manually recreate DB1's structure inside it would involve copying all the data, violating the 'data is NOT copied' requirement.