redis.md

Redis

Learning notes related to Redis data store

Redis is lightweight, persistent key value data storage

with flexible mechanism for organizing your data in terms of data structures

Redis is simple and performant

Usage

with Docker

$ docker pull redis

$ docker run redis

to enter container

$ docker exec -it <container_id> sh

Use cases

• Caching

  As value access in redis is super fast it's popular alternative for varnish, memcached.

• Real time data

  Redis often used for services that works with real time data. It has flexible mechanism for setting TTL for data. Due to exposing common data types you can achieve good perfomance when read-write your data.

Data types

Redis not a simply key-value storage like Memcached it is better called data structures service. Redis has set of common builtin data types each has own set of accepted commands. Most common one are String, Set, Hash, SortedSet

Common commands can be found here

Data modeling

• Mostly you store string key-value pairs. For storing complex json like structures you just use json.loads for stringifing it and store as pure string

• If you need more control over qeuring you can store objects like Hash structure. But be aware that Redis do not allow you to store nested Hashes. Probably you need recap you data relation model for such cases.

• For storing uniq uniq values you can use Set structure. Set has all common operations like union joins intersect , etc.

• Common way for storing separate records of data but related to same domain like Users or Twits is by prefixing each record with users:. So sample user object can be stored as a key users:{some_user_id}. For quering such data you can use keys command with pattern, but it's not a recommended way. Better way will be to store user as a hashes, but if you need strctly this structure you can create separate Set that will hold user related keys. So then querining operation will be O(x) where x is Set length instead of O(n) where n is global number of keys.

• There are lots of other recipes how to use Redis, but the beauty of Redis is that you can come with a way of how to use it it that nobody thought about before

Resouces

• Little Redis book
• Official Redis documentation
• aioredis for async Python client