PyPICloud - PyPI backed by S3

This is an implementation of the PyPI server for hosting your own python packages. It uses a three layer system for storing and serving files:

+---------+        +-------+        +-----------+
| Storage | <----> | Cache | <----> | Pypicloud |
+---------+        +-------+        +-----------+

The Storage layer is where the actual package files will be kept and served from. This can be S3 or a directory on the server running pypicloud.

The Cache layer stores information about which packages are in stored in Storage. This can be DynamoDB, Redis, or any SQL database.

The Pypicloud webserver itself is stateless, and you can have any number of them as long as they use the same Cache. (Scaling beyond a single cache requires some additional work.)

Pypicloud is designed to be easy to set up for small deploys, and easy to scale up when you need it. Go get started!

Code lives here: https://github.com/stevearc/pypicloud

Indices and tables