spark/internals
intermediate
Connecting…

The Spark DAG

The DAG is the graph of operators. Wide edges are shuffles. Reading the DAG is reading the cost.

Lesson 20 of 29 · Spark path

Explain it at my level

  1. Nodes
  2. Edges
  3. Shuffle edge
Watch the canvas:narrowshuffle / ExchangejoinLive simulation
FileScanFilterExchangewide edgeAggThis graph is the Spark DAG — not a Workflow.
DAG = operator graph. Wide edges cut stages. That is the Spark DAG.