Zappa VS starlette

If people aren't familiar, there's a similar project for Python that's fantastic: https://github.com/zappa/Zappa

Chalice was a consequence, a reaction from AWS to the release of (Zappa Framework)[https://github.com/zappa/Zappa] that provide a very good alternative to migrate very quickly a Django/Flask or any WSGI compliant solution in Python.

Use Zappa https://github.com/zappa/Zappa and host as a Lambda, simple setup and deployment, Lambda only costs when processing requests, no servers to mess around with

One of my favorite options that is probably the most cost-effective is to deploy using a 'severless' model on AWS Lambda using zappa which supports deploying Python webapps to AWS in this way. Zappa also makes it super easy to deploy in just a couple commands! The README includes instructions for everything you might need, including handling sensitive information like your database passwords, running django management commands, setting up DNS, etc.

Deployment works nicely with Docker. I often use AWS AppRunner because it's really easy and just scales. Some people use AWS Lambda with Zappa but I don't recommend it unless you really want to spend less than $15 a month. You will probably need Django Storages to save uploads to an S3 bucket. At some stage you might want to put a CloudFront distribution in front of everything but the configuration of the caching behaviour might be a bit confusing when you do it the first time.

It doesn't have to be an either-or! I have several apps in production that were developed on Django or Flask, and deployed to Lambda using Zappa.

If you need a relational DB, you can use AWS Aurora or RDS and use cloud functions ('lambda' in AWS) that you can invoke with HTTP to process the document first. Zappa will do a lot of the configuration for you if you go that route.

Lambda + API gateway, this library bundles a Django application into a lambda https://github.com/zappa/Zappa . 1 million free invokes from aws, scale to zero, plugs into your RDS

I use Zappa, it just schedules a frequent execution of the lambda: https://github.com/zappa/Zappa#keeping-the-server-warm

Fast Execution: FastAPI is built on top of Starlette and Pydantic, making it one of the fastest Python frameworks for building APIs.

The framework's efficiency comes from its use of Starlette for building asynchronous web services and Pydantic for robust data validation and serialization, powered by Python's type hints. Pydantic has recently announced the official release of Pydantic V2 (June 2023), which is a ground-up rewrite that offers many new features and performance improvements, so make sure to be using that instead of V1.

Uvicorn + Starlette 8k requests/s

Starlette documentation

Same here, but without these weird utils it doesn't get any better.

I have 7 YoE with Django. Its great at so many things. You see some code, like middlewares, and immediately understand what's going on.

Now, we also have Starlette. The base of all new, fancy asgi libraries. Here's the base middleware class.

https://github.com/encode/starlette/blob/8d7a1cacfb3e1a30cbb...

In the last couple of years I heard 'we're running fastapi on production. Wanna join us?' so many times... but the reality is that it's still not suitable for prod. Who wants to work with a code like that if you have a readable, stable Django? I'm clueless.

Set up Starlette, a tool we'll use to make async requests

I never used websockets in FastAPI so I wouldn't know how to guide you more, but Middleware in Websockets are 100% supported by Starlette : https://github.com/encode/starlette/issues/641

Websockets are the way but I would not recommend django as it's still not fully async. I would go for other tools.