DataYatriSpark, seen under the hood
Simulate

Simulation/Spark

Educational simulation
Spark

DataFrames & schemas

A distributed table with a schema — not an RDD of tuples, not pandas on the driver.

Watch the canvas:schemapartitioned rowsactionLive simulation
orders DataFrameorder_id: long · region: string · amount: decimalPartition 0executor-1Partition 1executor-2Partition 2executor-3filter(region == 'US').select(...)still a plan — no scan yetshow()action — few rows to driverNamed columns. This is the contract Catalyst will optimize.
DataFrame = distributed table + schema. That schema is why Catalyst exists.