event-fanout

Event Fanout

A single business event (e.g., "order placed") needs to trigger actions in multiple downstream systems simultaneously: update inventory, send a confirmation email, notify the warehouse, and log for analytics. Each consumer is independent and should not be blocked by a slow sibling.

Pipeline

[fo_receive_event]
     |
     v
     +───────────────────────────────────────────────────+
     | [fo_analytics] | [fo_storage] | [fo_notification] |
     +───────────────────────────────────────────────────+
     [join]
     |
     v
[fo_aggregate]

Workflow inputs: eventId, eventType, payload

Workers

AggregateWorker (task: fo_aggregate)

Aggregates results from the three parallel fan-out branches.

• Sets status = "all_completed"
• Reads analyticsResult, storageResult, notifyResult. Writes status, processorResults

AnalyticsWorker (task: fo_analytics)

Tracks event analytics and updates metrics.

• Sets result = "tracked"
• Reads eventId. Writes result, metricsUpdated

NotificationWorker (task: fo_notification)

Sends a notification about the event.

• Sets result = "notified"
• Reads eventId, eventType. Writes result, channel

ReceiveEventWorker (task: fo_receive_event)

Receives an incoming event and passes it through for fan-out processing.

• Reads eventId, eventType, payload. Writes eventId, eventType, payload

StorageWorker (task: fo_storage)

Stores the event payload to a data lake.

• Sets result = "stored"
• Reads eventId. Writes result, storageLocation

---

41 tests | Workflow: event_fanout_wf | Timeout: 120s

See RUNNING.md for setup and usage.