Basic: Parallel paths logic
This sample runs multiple paths in parallel and merges their output values.
Choose this when
Use this sample when independent pieces of work can run at the same time and the workflow should continue only after all branches finish.
What it demonstrates
- Splitting execution with FanOut.
- Waiting for branches with FanIn.
- Nesting a second FanOut and FanIn inside one branch.
- Merging values written under
output.
How it works
The workflow initializes shared context, fans out into parallel Code nodes, and each branch waits for a random short delay before writing its own output message.
One branch contains another fan-out and fan-in pair, showing that parallel orchestration can be nested.
The final FanIn waits for all branches and merges their output entries before the workflow ends.
Setup notes
No connector or model is required. The random delays make branch completion order visible in the editor.