Do I need a Spark cluster to practice?

No. The editor on each page runs your code in the browser, so you can focus on the DataFrame logic first.

What should I know before PySpark?

Basic Python and a feel for tables. SQL helps a lot, because many DataFrame operations have a direct SQL twin.

PySpark on LakeBench

Learn PySpark from the basics to big data

PySpark runs Python code on data too big for one machine. The track explains how Spark splits work, then teaches DataFrames, joins, windows and the habits that keep jobs fast. You practice in the browser without a cluster.

Start the first lesson Open PySpark 40

21
lessons
8
free lessons
40
PySpark sheet problems

What you will be able to do

After the sheet you can read and write DataFrame code, explain what happens when Spark runs it, and avoid the mistakes that make big jobs slow.

Learn it in order

PySpark from Basics to Big Data has 5 modules. 8 lessons are free.

Module 1
PySpark Architecture
5 lessons
Module 2
Mental model & DataFrames
3 lessons
Module 3
Aggregations, windows & nested data
6 lessons
Module 4
Performance at Scale
3 lessons
Module 5
Production Spark
4 lessons

See the full PySpark from Basics to Big Data track

Practice by topic

PySpark 40 lists 40 problems in a fixed order, 23 of them free. Work through one topic at a time.

Select, filter and add columns9
Count, sum and average by group9
Joins6
Arrays, text and reshaping4
Built-in functions instead of UDFs4
Window functions6
Deduplicate and keep the latest row2

Open the full sheet Try the first problem

PySpark topic guides

More for PySpark

Questions

Do I need a Spark cluster to practice?: No. The editor on each page runs your code in the browser, so you can focus on the DataFrame logic first.
What should I know before PySpark?: Basic Python and a feel for tables. SQL helps a lot, because many DataFrame operations have a direct SQL twin.