A distributed Redis cache backend for Django

• This cache backend requires the redis-py Python client library for communicating with the Redis server.
• This is a fork of django-redis-cache. Since a cache backend is not likely to change much, it doesn't make much sense to add more machinery and overhead just to allow a distributed cache setup. django-redis-cache should remain simple and fast.
• The sharding uses consistent hashing so you can add or remove nodes without resharding everything. However, since the server list is determined by the settings file, you can't dynamically add or delete cache servers. To do so in the future, there needs to be some kind of shared state between app servers. Probably a redis server to hold the configuration and status of each cache server. However, this will require a round trip from the app server to the cache configuration server before it can start the a normal cache lookup.
• Variable cache node consistent hashing requires a O(log n) lookup for every key access, where n is the number of cache servers. This is less ideal than the O(1) lookup we are used to, but necessary if you don't have a fixed power of 2 number of cache servers.

1. Run python setup.py install to install, or place dredis_cache on your Python path.
2. Modify your Django settings to use dredis_cache :

On Django >= 1.3:

CACHES = {
    'default': {
        'BACKEND': 'dredis_cache.RedisCache',
        'LOCATION': [
            '<host0>:<port0>',
            '<host1>:<port1>',
            '<host2>:<port2>',
            '<host3>:<port3>',
        ],
        'OPTIONS': { # optional
            'DB': 1,
            'PASSWORD': 'yadayada',
        },
    },
}

To run tests, you need four instances of a redis-server running on localhost, ports 6379-6382. Then, assuming redis-py is installed, type python runtests.py in the project root.