Skip to main content

Tornado architecture

Tornado, a Python web framework and asynchronous networking library, is a powerful tool for building high-performance web services. In this article, we will delve into the architecture of Tornado to better understand its workings.

1. Overview of Tornado Architecture​

Tornado's architecture is built to handle thousands of simultaneous connections, making it ideal for real-time web services. It uses a non-blocking network I/O and can scale to tens of thousands of open connections, making it perfect for long polling, WebSockets, and other applications that require a long-lived connection to each user.

1.1 Non-blocking I/O​

Tornado is based on non-blocking I/O (Input/Output) operations. Traditional web servers use blocking I/O, which means they wait for data before continuing their operations. Non-blocking I/O, on the other hand, allows Tornado to serve other requests while waiting for data, resulting in a highly efficient use of resources.

2. Event Loop​

The event loop is at the core of Tornado's architecture. It's a programming construct that waits for and dispatches events or messages in a program. In Tornado, it's provided by the IOLoop class. The IOLoop is responsible for handling all I/O events and dispatching them to their respective Handlers.

2.1 Handlers​

Handlers in Tornado are responsible for processing requests and producing responses. When a request arrives, a handler is created, and the appropriate method on that handler is invoked based on the type of HTTP request.

3. Tornado’s Asynchronous Libraries​

Tornado provides support for third-party libraries, allowing them to operate asynchronously. These libraries include AsyncHTTPClient (for non-blocking HTTP requests), AsyncTestCase and gen_test (for asynchronous unit tests), and many others.

4. Application and RequestHandler​

In Tornado, an Application is a collection of request handlers that make up a web application. The RequestHandler does the work of handling the request, forming the response, and deciding how to interact with the client.

4.1 Application​

The Application class is the heart of any Tornado web application. It's where your URL routes and settings are defined.

4.2 RequestHandler​

The RequestHandler class is where you define how to respond to any given request. Each HTTP verb (GET, POST, PUT, DELETE) corresponds to a method on your RequestHandler.

5. Templates and UI Modules​

Tornado also provides a template language. This allows you to generate HTML from Python code, which makes building dynamic web pages a breeze. UI Modules are a feature that allows for reusable components (like headers and footers) in your templates.

6. Conclusion​

That's a brief overview of Tornado's architecture. Tornado uses a non-blocking network I/O and an event loop to handle thousands of simultaneous connections. It's a powerful framework that's perfect for real-time web services.

Stay curious and keep exploring Tornado's amazing capabilities!