While using Redis one must take care of
* Select a consistent method to name and prefix your keys. Manage your namespace
* Create a “Registry” of key prefixes that maps each of your internal documents for that application which “own” them
* For every class you put through into your Redis infrastructure: design, implement and test the mechanisms for garbage collection or data migration to archival storage
* Design, implement and test a sharding library before you have invested much into your application deployment and make sure that you keep a registry of “shards “replicated on each server
* Separate all your K/V store and related operations into your own library/API or service
