Sensoworks Edge/Fog architecture

Overview

The Sensoworks Fog architecture is designed to manage configurations, control flows, and monitor the status of various processes in a distributed environment. This document provides an overview of the architecture, including the main components and their interactions.

Components

1. Configuration Server

The ConfigurationServer is a Flask-based server that exposes RESTful APIs for managing configurations and controlling flows. It interacts with the ConfigManager and FlowManager to handle configuration data and flow commands.

2. ConfigManager

The ConfigManager handles the storage, retrieval, and manipulation of configuration data. It is responsible for reading configuration files, updating configurations, and providing configuration details to other components.

3. FlowManager

The FlowManager is responsible for managing the lifecycle of flows. It can start, stop, pause, and resume flows based on the commands received from the ConfigurationServer. The FlowManager maintains the status of each flow and interacts with the SensoworksFog instances to control their execution.

4.Workflow Engine (SensoworksFog class)

The SensoworksFog class represents a single flow instance and acts as the workflow engine. It is initialized with a configuration and is responsible for managing the execution of tasks within the flow. The SensoworksFog class handles triggering tasks based on defined conditions, managing state transitions, and ensuring tasks are executed in the correct order. Each SensoworksFog instance runs in a separate thread to allow concurrent execution of multiple flows.

5.Flows

The Flows component provides a user interface for managing configurations that the workflow engine will execute. This interface enabling users to efficiently set up and control their workflows within the IoT infrastructure. In addition to its configuration management capabilities, the Flows component communicates with the Edge-Broker component to exchange and synchronize information with the platform. This communication ensures that all configurations and workflow instructions are up-to-date and accurately reflect the current state of the system. Furthermore, the Flows component provides detailed information about the node on which it resides. This information is critical for maintaining an accurate and comprehensive view of the network's operational status, enhancing the platform's ability to manage and optimize its resources. By integrating with the Edge-Broker and providing a robust interface for workflow management, the Flows component plays a crucial role in the seamless operation of the IoT system, ensuring that configurations are effectively managed and executed across the network.

Interactions

Initialization

1. MainSensoworksFogGateway: This is the main entry point for the application. It initializes the ConfigManager, FlowManager, and ConfigurationServer, and starts the server and WorkflowEngine instances as background processing threads.
2. ConfigurationServer: The server starts and listens for incoming API requests.

Configuration Management

1. Get All Configurations: The ConfigurationServer receives a request to retrieve all configurations. It calls the get_all_configs method on the ConfigManager, which reads and returns all configuration files.
2. Get Configuration by ID: The ConfigurationServer receives a request to retrieve a specific configuration. It calls the get_config method on the ConfigManager with the flow_id, and the ConfigManager returns the corresponding configuration.
3. Add/Update/Delete Configuration: The ConfigurationServer handles requests to add, update, or delete configurations by calling the respective methods on the ConfigManager.

Flow Control

1. Start Flow: The ConfigurationServer receives a request to start flows. It calls the start_flows method on the FlowManager with the list of flow_ids. The FlowManager initializes and starts a SensoworksFog instance for each flow.
2. Stop/Pause/Resume Flow: Similar to starting flows, the ConfigurationServer handles requests to stop, pause, or resume flows by calling the respective methods on the FlowManager.

Status Monitoring

1. Get Flow Status: The ConfigurationServer receives a request to retrieve the status of all flows or a specific flow. It calls the get_flow_status method on the FlowManager, which returns the current status information.
2. Current Task: The FlowManager can also provide details about the current task being executed in a flow.

Data Flow

1. Configuration Files: Stored in a specified directory and managed by the ConfigManager.
2. Flow Status: Maintained in a shared dictionary with thread-safe access managed by the flow_status_lock.
3. Commands and Responses: Handled by the ConfigurationServer and passed to the FlowManager for execution.

Example Workflow

1. Starting the Server: The MainSensoworksFogGateway starts the ConfigurationServer and the background engine flow.
2. Adding a Configuration: A client sends a POST request to /flows with a new configuration. The ConfigurationServer processes the request and updates the configuration files via the ConfigManager.
3. Starting a Flow: A client sends a PATCH request to /flows/commands with a start command and flow IDs. The ConfigurationServer forwards the command to the FlowManager, which initializes and starts the SensoworksFog instances.
4. Monitoring Status: A client sends a GET request to /flows/status to retrieve the status of all flows. The ConfigurationServer fetches the status from the FlowManager and returns it to the client.