litecache

Introduction

---

litecache is a basic in-memory key-value store written in Python supporting a subset of RESP (redis serialization language)

Features

---

• Key-Value Cache: Store and retrieve data using a simple key-value pair system
• RESP: support a subset of resp so should work with any redis client (SET,GET,DEL)
• Simple: small code base and fairly readable so should be easy to understand and make updates to

Installation

---

To install litecache, run the following command:

pip install litecache2

or

pip install git+https://github.com/psqnt/litecache.git

Usage

---

Basic Example

$ litecache --host localhost --port 6399

(development)

$ uv run litecache

using docker:

$ docker build -t litecache:latest .
$ docker run litecache:latest

Configuration

---

litecache can be configured using the following cli parameters:

Usage: litecache [OPTIONS]

  Start the LiteCache async TCP server.

Options:
  --config TEXT   Path to config YAML file
  --host TEXT     Host to bind to
  --port INTEGER  Port to listen on
  -v, --verbose   Verbosity level (e.g., -v for INFO, -vv for DEBUG)
  --help          Show this message and exit.

Environment Variables

env vars take precedence

export LITECACHE_HOST=0.0.0.0
export LITECACHE_PORT=9999
export LITECACHE_VERBOSE=1

Configuration File

host: "0.0.0.0"
port: 6409
verbose: 1

uv run litecache --config /path/to/config/filename.yml

Testing

---

To run the tests, use the following command:

uv run pytest

Redis-cli and Redis Benchmark

redis-cli

➜  ~ redis-cli SET foo bar
OK
➜  ~ redis-cli GET foo
"bar"

redis-benchmark

➜  ~ redis-benchmark -t set,get,del -n 100000 -c 50 -q

SET: 32509.75 requests per second, p50=1.343 msec
GET: 36258.16 requests per second, p50=1.191 msec