Multi Layer Cache

[rspadim]

Hi folks!

Could be possible implement a multi layer cache? example:

0. creating the cache:

'cache': [
    {
        'cache': "aiocache.SimpleMemoryCache",
        'serializer': {
            'class': "aiocache.serializers.StringSerializer"
        }
    },
    {
        'cache': "aiocache.RedisCache",
        'endpoint': "127.0.0.1",
        'port': 6379,
        'timeout': 1,
        'serializer': {
            'class': "aiocache.serializers.PickleSerializer"
        }
    }
]

1. cache.set("key", "value") / cache.delete("key")
   1.1) cache will execute function to all layers

2. cache.get("key")
   2.1) a for i in layers: will check what's the first cache that have this data
   2.2) if it's not the first layer and data was found, update from current layer to first layer

why? the idea is have a memory cache with less cached items than the redis cache, and when it's not found on local memory check redis value

[argaen]

Hey @rspadim yeah this is a common use case that I see repeated in different places but it is a completely new dimension of implementation and feature. Feel free to add a proposal PR with the feature and we can discuss there! We could name the new type something like "MultiLayerCache" or something like that.

However, to cover the case you mention about having less data in memory cache you would need something that helps limit that #498