Redis Server in C

This ongoing project is a lightweight Redis compatible server implementation in C, created as part of the Coding Challenges project.

Latest benchmark results

Features

• RESP (REdis Serialization Protocol) implementation
• Support for the following Redis commands (PING, ECHO, SET (with options), GET, EXIST, DELETE, INCR, DECR, LPUSH, RPUSH)
• GoogleTest for testing (unit tests and integration tests)
• CMake for building

Technical Implementation

• epoll-based I/O multiplexing
• Ring buffers (input and output) for each client
• Zero allocation byte parsing for RESP protocol

System Requirements

• Linux operating system (or other Unix-like systems)
• CMake (version 3.10 or higher)cmake
• GCC or Clang compiler with C11 support

Build Instructions

This project uses CMake for building.

mkdir build
cmake -S . -B build
cmake --build build

Running the server

After building the project, you can run the server with the following command

./build/redis-lite

The server will start listening on port 6379 by default.

Once the server is running, you can use redis-cli to test various commands.

Start the redis-cli, and connect to the server.

redis-cli
127.0.0.1:6379> PING
PONG
127.0.0.1:6379> ECHO "Hello World"
"Hello World"
127.0.0.1:6379> SET key value
OK
127.0.0.1:6379> GET key
"value"

Using SET with Options

The SET command in this Redis server implementation supports additional options for enhanced functionality:

The syntax is:

SET key value [NX | XX] [GET] [EX seconds | PX milliseconds |
  EXAT unix-time-seconds | PXAT unix-time-milliseconds | KEEPTTL]

Key Expiry

You can set an expiration time for a key using various key expiration options:

• EX: Set the expiration time in seconds.
• PX: Set the expiration time in milliseconds.
• EXAT: Set the expiration time to a specific Unix timestamp (in seconds).
• PXAT: Set the expiration time to a specific Unix timestamp (in milliseconds).
• KEEPTTL: Retain the existing TTL (time-to-live) of the key.

Using GET Option

The GET option allows you to return the old value stored at the key while setting a new value. If the key does not exist, it will return nil.

Using EX:

127.0.0.1:6379> SET mykey "value" EX 60  # Key expires after 60 seconds
OK

Using PX:

127.0.0.1:6379> SET anotherkey "value1" PX 500  # Key expires after 500 milliseconds
OK

Using EXAT:

127.0.0.1:6379> SET mykey2 "value2" EXAT 1672531199  # Key expires at specific Unix timestamp
OK

Using PXAT:

127.0.0.1:6379> SET mykey3 "value3" PXAT 1672531199000  # Key expires at specific Unix timestamp in milliseconds
OK

Using KEEPTTL:

127.0.0.1:6379> SET mykey4 "value4" KEEPTTL  # Retain the existing TTL of mykey4
OK

Using GET:

127.0.0.1:6379> SET cool_key cool_value GET # SET cool_key to cool_value and return previous value if it existed
nil # Since cool_key did not exist before, the previous value is nil.
127.0.0.1:6379> SET cool_key another_value # SET cool_key to cool_value and return previous value if it existed
cool_value

Conditional Setting with NX and XX

NX: Only set the key if it does not already exist. XX: Only set the key if it already exists.

Running Tests

After building the project, you can run the tests using CTest, CMake's testing tool. Follow these steps:

cd build
ctest