Reading a Physical Plan

All four phases end in a physical plan, and being able to read it is the practical payoff of understanding the phases. The plan is a tree of operators, printed by explain, and you read it from the bottom up because that is the order data flows: leaves are scans, and each operator consumes the output of the one below. A handful of operators and markers carry most of the meaning, and once you know them the plan reads as a diagnosis. The first thing to do with any plan is count the Exchange nodes, because each one is a shuffle and shuffles are where the cost lives. A plan with one Exchange shuffled once; three Exchanges means three reorganisations of the data. The second thing is to check the join operators: a BroadcastExchange feeding a BroadcastHashJoin tells you Catalyst chose to broadcast

About This Interactive Section

This section is part of the Inside Catalyst: The Four Phases lesson on DataDriven, a free data engineering interview prep platform. Each section includes explanations, worked examples, and hands-on code challenges that execute in real time. SQL queries run against a live PostgreSQL database. Python runs in a sandboxed Docker container. Data modeling problems validate against interactive schema canvases. All content is framed around what data engineering interviewers actually test at companies like Meta, Google, Amazon, Netflix, Stripe, and Databricks.

How DataDriven Lessons Work

DataDriven combines four interview rounds (SQL, Python, Data Modeling, Pipeline Architecture) with adaptive difficulty and spaced repetition. Easy problems get harder as you improve. Weak concepts resurface until you master them. Your readiness score tracks progress across every topic interviewers test. Every lesson section ends with problems you solve by writing and running real code, not by picking multiple-choice answers.