parallel

Parallel Workflow Example

This example demonstrates how to define parallel execution branches in a workflow using the Parallel builder method.

Overview

The workflow splits execution into two parallel branches that process the same input independently.

It showcases:

• Parallel Definition: Using .Parallel() to define steps that branch off from the previous step.
• Independent Processing: Steps receiving the same input and processing it differently.

Workflow Structure

1. Start Step (start):

   • Input: WorkflowInput
   • Action: Logs start of workflow.
   • Output: WorkflowInput (Pass-through)

2. Parallel Branch: The workflow splits into two steps executing conceptually in parallel:

   • Add Step (add):

     • Input: WorkflowInput
     • Action: Adds Val1 + Val2.
     • Output: AddOutput

   • Multiply Step (multiply):

     • Input: WorkflowInput
     • Action: Multiplies Val1 * Mult.
     • Output: MultiplyOutput

3. Format Step (format):

   • Input: MultiplyOutput (Receives output from the last executed parallel step)
   • Action: Formats the result.
   • Output: FormatOutput

Data Flow

graph TD
    Input[WorkflowInput] --> Start[Start Step]
    Start --> Add[Add Step]
    Start --> Multiply[Multiply Step]
    Multiply --> Format[Format Step]
    Format --> Output[FormatOutput]

Loading

Note: In the current engine implementation, parallel steps are executed in a deterministic topological order. The step following a parallel block receives the output of the last executed step in that block (in this case, Multiply).

Running the Example

1. Start the Server:

   go run main/main.go

2. Trigger the Workflow:

   curl -X POST http://localhost:3000/api/v1/workflows/simple-math \
     -H "Content-Type: application/json" \
     -d '{"val1": 10, "val2": 5, "mult": 2}'

Key Code Concepts

Parallel Builder

The Parallel method adds multiple steps that all branch from the previous step(s).

builder.
    Sequence(NewStartStep()).
    Parallel(
        NewAddStep(),
        NewMultiplyStep(),
    )