Free DP-900 practice test: 30 questions with answers

Azure SQL Database is a managed relational service that handles patching, backups and high availability. Cosmos DB is a multi-model NoSQL service, Blob Storage stores objects, and Redis is an in-memory cache.

SQL Server on an Azure VM gives full control over the operating system and engine. Cosmos DB and Table storage are non-relational, and Data Lake Storage holds files for analytics.

Geo-replication creates secondary replicas that can serve read traffic and support recovery. Backup retention, public IPs and text files do not provide replicated read scale.

Azure Database for PostgreSQL is a managed PostgreSQL service. Table storage is NoSQL, Synapse is an analytics platform, and file shares store files.

GROUP BY groups rows by region so an aggregate such as SUM can be computed for each group. ORDER BY sorts results, WHERE filters rows, and INSERT adds data.

An index on the filtered column speeds lookups. Triggers run code on changes, views save queries, and deletion procedures do not improve query speed.

Point-in-time restore recovers a database to a moment within the retention window. Public IP restrictions, read replicas and laptop exports do not restore a past state.

Transparent data encryption encrypts data at rest and can use customer-managed keys in Key Vault. Disabling encryption, storing keys in the same table and encrypting headers do not meet the requirement.

Structured data follows a fixed schema with defined columns, as in relational tables. Semi-structured data such as JSON has flexible fields, and unstructured data such as images has no schema.

JSON documents carry their own schema information but vary in fields, which makes them semi-structured. A fixed relational schema would be structured, and binary files would be unstructured.

Normalization organizes data into related tables that reduce duplication and protect consistency. Replication copies data, sharding partitions it, and caching stores frequently used results.

Atomicity ensures a transaction is all or nothing. Latency is response time, partitioning splits data, and elasticity relates to scaling resources.

A primary key must uniquely identify each row. It does not need to change frequently, store multiple values, or be text.

A foreign key references a primary key in another table, establishing a relationship. An index speeds lookups, a view is a saved query, and a trigger runs code on events.

Streaming workloads process continuous data as it arrives. Batch workloads process collected data at intervals, transactional workloads focus on individual updates, and archive retrieval is not a query pattern for live streams.

ETL or ELT describes extracting, transforming and loading data for analysis. Encryption, network segmentation and load balancing address other concerns.

Cosmos DB is a globally distributed multi-model database with low-latency reads and flexible documents. SQL Database is relational, Data Factory orchestrates data movement, and Monitor collects telemetry.

Blob Storage stores unstructured objects such as images and video and provides HTTP access. SQL Database and Synapse are not object stores, and Cosmos DB is not for relational queries.

Key-value, document and column-family models are non-relational. Relational tables and star schemas are relational or dimensional designs.

Azure Cache for Redis provides in-memory key-value storage with expiration and fast responses. SQL Database, Data Lake Storage and Files are not designed for millisecond session caches.

Azure Table storage is a NoSQL key-value service for large amounts of structured, non-relational data with low cost. SQL Managed Instance is relational, Synapse is for analytics, and Machine Learning is for model development.

Cosmos DB for NoSQL stores JSON documents and supports a SQL-like query language. Blob Storage lacks document queries, Azure Files is file storage, and Monitor collects telemetry.

Cosmos DB supports graph data through the Apache Gremlin API for traversing relationships. Blob, Table and Files do not provide graph traversal.

Synapse Analytics supports large-scale analytics and Power BI provides reports and dashboards. Redis, Files and Front Door are not analytics engines.

Azure Synapse Analytics combines warehousing, big data and integrated analytics with serverless SQL and notebooks. The other services do not provide this integrated analytics platform.

Power BI creates interactive dashboards and reports from data models and shares them. The other services do not provide dashboards.

Azure Data Factory orchestrates scheduled data movement and transformation with run monitoring. Firewall, Bastion and DNS do not move data.

Data warehouses are optimized for large analytical queries over consistent historical data. They are not designed for many small updates, do not replace operational systems, and support aggregation.

Synapse Spark pools and Azure Databricks run Spark for big data processing with notebooks. Bastion, DNS and Storage Explorer do not run Spark.

Ingesting events, processing them in Stream Analytics and streaming results into Power BI supports real-time dashboards. Static PDFs, printed receipts and manual spreadsheets are not real-time.