Data Flows

Data Flows

1. Chain Initialization

• Process:
  • The NodeSupervisor configures and distributes nodes.
  • Nodes request ReportingAgent instances from the MonitoringAgent.
• Flow:

  NodeSupervisor → Create Chain Config → Distribute Nodes → Set Up Monitoring
  

2. Execution Flow

• Process:
  • Nodes process incoming data via PipelineProcessors.
  • Status updates are continuously sent via the ReportingAgent.
  • Processed data is forwarded to the next node in the chain.
• Flow:

  Node → Process Data → Report Status → Forward Data
  

3. Monitoring Flow

• Process:
  • ReportingAgents capture status changes from each node.
  • The MonitoringAgent aggregates these updates and broadcasts chain status externally.
• Flow:

  ReportingAgent → Capture Status → MonitoringAgent → Broadcast Status
  

---

Typical Workflow Example

1. Configure Callbacks: Set up service and monitoring callbacks.
2. Create Chain Configuration: Define nodes, services, and locations.
3. Deploy Chain: Issue a CHAIN_DEPLOY signal via NodeSupervisor.handleRequest.
4. Prepare Distribution: The NodeSupervisor sets up local/remote nodes.
5. Start Chain: Initiate execution with a CHAIN_START signal and provide input data.
6. Execute and Monitor: Nodes process data while the MonitoringAgent tracks progress and aggregates status updates.

---

Error Handling and Logging

• Mechanism: A built-in logging system (Logger) tracks operations, warnings, and errors.
• Customization: Users can configure logging outputs as needed, ensuring that error handling aligns with specific workflow requirements.