Skip to main content

How Operations Work​

To understand how operations function, it helps to consider the underlying technologies and concepts they leverage. Basic familiarity with Docker and Python is recommended before proceeding.

Operations serve as a bridge between external tools and the Stage² platform by containerizing these tools and wrapping them in a consistent, standardized interface. This approach forms the foundation of Stage²’s tool-agnostic design—enabling users to integrate, automate, and scale sophisticated workflows without being tied to a specific technology stack.

Note: Include a graphic illustrating containerizing a tool and organizing those containers into a workflow.

Key Components of an Operation​

1. Containerization​

Each operation runs inside a Docker container. This container includes:

  • The software or tool the operation will execute.
  • All necessary dependencies to ensure a stable, consistent environment.
  • A custom interface script—stage2_interface.py—which acts as the operation’s entry point.

By encapsulating each tool in a container, operations remain portable and reliable, irrespective of the underlying infrastructure.

2. Standardized Inputs and Outputs​

Operations in Stage² adhere to a standardized input-output model:

  • Inputs: Operations receive assets (e.g., files, datasets) and parameters (e.g., settings, configurations) in a structured JSON format. This format defines all the information required to execute the tool effectively.
  • Outputs: The results of the operation—such as processed files, data, or visualizations—are generated and stored in a specified location, ready for downstream operations in the workflow.

This standardization simplifies the integration of diverse tools and ensures that each operation can interact smoothly with others.

3. Stage² Interface Script​

At the heart of each operation is a interface script that can be in any programming language, example in python stage2_interface.py. This script is responsible for:

  • Parsing and validating the inputs: It reads the structured JSON input provided by Stage², ensuring all required parameters and assets are present and correctly formatted.
  • Executing the tool: It runs the underlying tool or software within the container, applying the provided parameters to generate the desired output.
  • Producing standardized outputs: After execution, it places the results in a known output directory or file, preparing them for subsequent operations in the workflow.

This script acts as the glue that binds the containerized tool to the Stage² platform, ensuring a consistent execution pattern across all operations.

Orchestration by Stage²​

Once you define and containerize your operation, Stage² takes care of:

  1. Invocation:
    Stage² launches the operation’s Docker container and provides it with the required inputs (see operation modal).

  2. Monitoring:
    Stage² tracks the operation’s execution, ensuring it completes successfully or handling errors if they occur.

  3. Routing Outputs:
    Once the operation finishes, Stage² makes the operation’s outputs available to other nodes in the workflow graph, enabling seamless data flow between operations.