Overview

Redis is an in-memory data structure store that goes beyond the conventional role of a database, extending its capabilities to serve as a cache and a message broker. It is designed to handle data with unparalleled speed and efficiency, having the capability to perform 110,000 SETs and 81000 GETs per second.

As a NoSQL key-value database, Redis simplifies data access through a single or set of keys to retrieve or delete associated values. In contrast to traditional relational NoSQL databases, Redis uses key-value pairs, where keys can be simple strings, serving as unique identifiers tied to specific data locations.

Redis provides a rich array of data structures such as strings, hashes, lists, sets, sorted sets, bitmaps, hyperloglogs, or geospatial indexes. This versatility empowers developers to mold their data storage according to specific application requirements, allowing for seamless integration with a wide range of use cases.

Redis Use Cases

Session Data
Redis serves as a session store for sharing session data among stateless servers in a web application. When a user logs in, session data, including a unique session ID, is stored in Redis. The session ID is returned to the client as a cookie. During subsequent requests, the client includes the session ID, allowing stateless web servers to retrieve the session data from Redis. It's crucial to note that Redis is an in-memory database, and session data stored in Redis will be lost if the Redis server restarts.

Replication
In addition to persistence options like snapshots and AOF, Redis uses replication for session data backup. While options like snapshots and AOF may be slow to load on restart, replication involves duplicating data to a backup instance. In the event of a main instance crash, the backup is swiftly promoted to handle traffic, ensuring quick recovery in production scenarios.

Distributed Log
Redis can be utilized as a distributed lock for coordinating access to shared resources among multiple nodes in an application. Clients attempt to acquire the lock by setting a key with a unique value and a timeout. If successful, the lock is acquired; otherwise, the client retries until the lock is released. While a basic implementation may lack full fault tolerance, various Redis client libraries offer high-quality distributed lock implementations out of the box for production use.

Ray Limiter
Redis can function as a rate limiter by utilizing its increment command on counters with set expiration times. In a basic rate-limiting algorithm, each incoming request uses the request IP or user ID as a key, and the request count is incremented using Redis's INCR command. Requests exceeding the rate limit are rejected. Keys expire after a specified time window, resetting counts for the next window. Redis supports more advanced rate-limiting algorithms, such as the leaky bucket algorithm, offering flexibility in implementing sophisticated rate limiters.

Gaming Leaderboard
For gaming leaderboards, especially in games of moderate scale, Redis is a preferred choice. The implementation relies on Sorted Sets, a fundamental Redis data structure. Sorted Sets consist of unique elements, each with an associated score, and they are sorted by score. This structure enables efficient retrieval of elements by score in logarithmic time, making Redis a delightful solution for various types of gaming leaderboards.

Getting hands-on with Redis

You can connect to Redis in the following ways:

1. With the redis-cli command line tool
2. Use RedisInsight as a graphical user interface
3. Via a client library for your programming language

Let's go over some of the basic commands in Redis to familiarize ourselves with how Redis works

Set and get a key-value pair
SET foo 100
This sets a key value-pair in the database with foo as the key and 100 as the value
To get this value, we can simply use
GET foo
This will return the value 100

Check if a key exists
We can use the EXISTS command to check if a key exists in the database or not. It will return 1 if the key exists and 0 otherwise EXISTS foo will return 1 whereas EXISTS foo1 will return 0

Deleting a key value pair
DEL command deletes the key from the database. DEL foo will return 1 indicating foo has been deleted. EXISTS foo will now return 0

Setting multiple key value pairs
The MSET command in Redis allows you to set multiple key-value pairs in a single command.
MSET key1 "value1" key2 "value2"
In this example, the MSET command sets the values for two keys (key1, key2) in a single go. After executing this command, key1 will have the value "value1" and key2 will have the value "value2"

Appending to a string
The APPEND command allows you to append a value to an existing string. For instance, say you have a key value pair as bar: "Hello" so APPEND bar " World" will result in the value of 'bar' becoming "Hello World".

List operations
Redis supports lists. You can use the LPUSH and RPUSH commands to push elements to the left and right of a list, respectively.
LPUSH myList "value1"
LPUSH myList "value2"
LPUSH myList "value3"
After executing these commands, the myList will contain the values in the following order:
"value3" "value2" "value1"
To retrieve all elements from the list:
LRANGE myList 0 -1
The LRANGE command with the range 0 -1 will return all elements of the list.

Set operations
Redis supports sets. Sets can be manipulated using commands like SADD to add elements and SMEMBERS to retrieve all elements of a set.
SADD mySet "value1"
SADD mySet "value2"
SADD mySet "value3"
After executing these commands, the mySet will contain the elements: "value1" "value2" "value3"
To retrieve all elements from the set:
SMEMBERS mySet
The SMEMBERS command will return all elements of the set.

Sorted Sets
Redis also supports Sorted Sets, a data structure that extends the functionality of sets by associating each element with a score. Commands like ZADD are used to add elements to a Sorted Set along with their scores. For example:
ZADD mySortedSet 1 "value1"
ZADD mySortedSet 2 "value2"
ZADD mySortedSet 3 "value3"
After executing these commands, the mySortedSet will contain elements with associated scores: "value1" (score 1), "value2" (score 2), and "value3" (score 3).
To retrieve elements from a Sorted Set based on their scores, you can use commands like ZRANGE or ZREVRANGE:
ZRANGE mySortedSet 0 -1
This command will return all elements of the Sorted Set in ascending order of their scores. You can adjust the range parameters for more specific retrievals.
Sorted Sets in Redis provide a way to maintain an ordered collection of unique elements, allowing for various operations based on element scores.

These examples cover a range of basic Redis commands, showcasing its versatility in handling different data structures and operations.

Redis, with its speed, versatility, and simplicity, invites you to explore the vast possibilities it unfolds. Whether you're building a real-time application, optimizing data access, or architecting a scalable solution, Redis stands ready to elevate your data management experience.