DP-900 Describe core data concepts Practice Question
Which TWO of the following are correct descriptions of data processing workloads in Azure?
⚠ Common exam trap
A common mix-up: candidates confuse 'streaming' with 'interactive querying' or assume batch can handle real-time data, but Azure explicitly separates these workloads based on data state (in motion vs. at rest) and latency requirements.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Streaming processing is used to process data in real time as it arrives.
Streaming processing in Azure (e.g., Azure Stream Analytics, Event Hubs, or Kafka on HDInsight) is designed to ingest, analyze, and act on data in near real-time as it arrives, often with sub-second latency. This is fundamentally different from batch processing, which handles data at rest.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Streaming processing is used for interactive queries on historical data.
Why it's wrong here
Interactive queries on historical data rely on query engines that can scan stored datasets and return ad-hoc answers, which is fundamentally an analytical or batch workload, not a streaming one. Streaming processing reads events as they occur from sources like IoT hubs or event brokers and applies computations in motion, so it has no concept of querying archived history on demand. Azure Stream Analytics, for example, uses temporal windows over live input, whereas interactive historical analysis is suited to Azure Synapse Serverless or Azure Databricks SQL on data at rest.
- ✗
Streaming processing is used to process data at rest.
Why it's wrong here
Data at rest refers to data persisted in storage such as Azure Blob, Data Lake, or database tables, and workload to process it is inherently batch or interactive analytics, not streaming. Streaming processing consumes data in motion—continuous, real-time event flows—and applies operations before or as the data is stored, not to previously stored files. A common misconception is that streaming can retroactively process historical data, but streaming engines like Azure Stream Analytics operate only on live windows of incoming events and cannot begin processing data that is already at rest.
- ✓
Streaming processing is used to process data in real time as it arrives.
Why this is correct
Streaming processing is purpose-built for real-time data: it ingests events continuously from sources like Azure Event Hubs or IoT Hub and processes them as they arrive, often with sub-second latency. This architecture enables real-time dashboards, anomaly alerts, and event-driven responses where decisions must be made on the latest data. For example, a streaming pipeline might aggregate clickstream events into 5-second windows to show current user activity, which is impossible with batch processing that defers computation until a scheduled run.
- ✗
Batch processing is used to process data in real time as it arrives.
Why it's wrong here
Batch processing inherently introduces latency because it first accumulates data over a fixed interval (e.g., hourly or daily) before launching jobs, so it cannot react to events as they arrive. Real-time processing requires an event-driven, continuous execution model that processes each record or micro-batch with minimal delay, which is the domain of streaming engines like Azure Stream Analytics. Running a batch job on a schedule means data will be analyzed only after the batch boundary, not the moment it is generated, so the statement conflates batch's delayed, bounded datasets with streaming's unbounded, low-latency paradigm.
- ✓
Batch processing is used to process large volumes of data at scheduled intervals.
Why this is correct
Batch processing is designed to efficiently process large, bounded datasets by grouping input into self-contained jobs that run at predefined times, such as nightly ELT cycles or monthly metric aggregations. This model is ideal for high-throughput workloads where latency of minutes to hours is acceptable, and it leverages distributed compute frameworks like Azure Databricks or HDInsight to scale out across many nodes. Because the entire dataset is known before the job starts, batch can perform comprehensive transformations, deduplication, and sum-of-parts calculations reliably, making it the standard for populating data warehouses and data lakes.
Go deeper
Related to this question
Learn chapter
Data Roles and Core Concepts
Key term
Batch processing
Batch processing is a method of running high-volume, repetitive data jobs where a group of transactions is collected, processed together automatically, and then results are produced without real-time user interaction.
Key term
Data
Data is raw, unprocessed information, like numbers, words, or measurements, that can be stored, processed, and analyzed by computers.
About these practice questions
Courseiva writes every DP-900 question from scratch — 820 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This DP-900 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the DP-900 exam.