Describe core data concepts Study Guide — DP-900

29% of the exam · domain 1 of 4 · Azure Data Fundamentals · Free Describe core data concepts practice questions

0 of 25 topics covered
What the exam asks here — tap to tick off0 / 7 covered
Domain 1 of 4

Describe Core Data Concepts – DP-900 Study Guide

This domain accounts for 25–30% of the exam and lays the vocabulary everything else builds on: how data is represented, how it's stored, the two great workload families (transactional vs analytical), and who does what on a data team. Questions here are conceptual — they describe a situation and ask you to name the correct category. Nail the definitions and the distinctions and this is the most reliable-scoring domain on the exam.


Ways to Represent Data

Three classifications, and the exam tests your ability to place a scenario in the right one:

TypeDefinitionExamples
StructuredFixed schema; every entity has the same fields; naturally tabular (rows/columns)Relational tables, a customer table with fixed columns
Semi-structuredHas organization (keys, nesting) but records may vary in shapeJSON, XML, key-value with flexible attributes
UnstructuredNo enforced internal structureImages, video, audio, PDFs, free-form text

The tell: "every record has exactly the same fields" → structured. "JSON with optional/nested fields that differ per record" → semi-structured. "media files / documents" → unstructured.


Common Data File Formats

Recognize each format's shape and best use:

  • CSV — simple, row-based, comma-separated text. Human-readable, universal, but no types or compression.
  • JSON — semi-structured, nested key-value. The web/API lingua franca; great for flexible documents.
  • XML — verbose hierarchical text with tags. Older interchange format.
  • Parquet — binary, columnar. Reads only the columns a query needs, with strong compression — the analytics favorite for large tables.
  • Avro — binary, row-based with schema embedded. Efficient for write-heavy and streaming pipelines.

Exam trap: "wide table, query reads a few columns, best performance" → Parquet (columnar). Don't pick CSV for performance — its simplicity doesn't help query speed.


Sign in free to read the full guide

5 more sections (about 497 words) continue below. The first domain's study guide is free for every DP-900 learner with a free account.