Memify is a lightweight, high-performance in-memory data store implemented in C++. Inspired by Redis, Memify provides efficient key-value storage with additional features like HMAC signature verification for secure communications. This project was built from the ground up, including features such as chunk-based processing for large data handling and an LRU (Least Recently Used) caching mechanism. The idea is to add more accurate geo-spatial and time series functions.
-
High-Performance Key-Value Storage: Optimized for speed, made using C++.
-
Security: Implements HMAC signature verification on both the server and client to ensure security.
-
Chunk-Based Processing: Handles large data transfers by processing data in manageable chunks.
-
LRU Cache: Built-in support for Least Recently Used caching to manage memory and enhance performance.
To build and run Memify, you will need:
-
A C++ compiler that supports C++11 or later (e.g., GCC, Clang)
-
OpenSSL library for secure key generation
-
CMake for building the project
- Clone the Repository
git clone https://github.com/Soumyadas15/memify.git
cd memify
- Generate a Secure 32-Byte Key Using OpenSSL
Memify requires a secure key for HMAC signature verification. Generate a 32-byte key using OpenSSL:
openssl rand -base64 32
Copy the generated key, which will be used in the config.ini file.
- Create a
config.iniFile
Create a config.ini file in the root directory of the project and add the following configuration:
[settings]
secret_key = "your-generated-32-byte-key"
port = "8080"
Replace "your-generated-32-byte-key" with the key generated in the previous step.
- Build the Project
mkdir build
cd build
cmake ..
make
- Run the Server
./Memify
The config.ini file is used to configure the server settings and security options. Ensure the paths to your TLS certificate and key files are correct, and update the secret_key with the secure key generated earlier.
To interact with Memify, you can use any framework of your choice. In your native environment, it will run on 127.0.0.1
As of now, Memify can process key-value data and Geospatial data. Also currently, you need to pass this data through a serializer.
Memify can act like a key-value in memory database. You can set and get values by providing their keys. To set keys:
SET <key> <value> <duration>Example:
SET name Soumya 200This will create a key called 'name' and set it's value to Soumya. The data will be stored for 200 seconds.
Similarly to find a key,
GET <key>Similarly to delete a key,
DELETE <key>Memify can act like a in memory database to compute geo-spatial data.
GEOSET <key> <value> <latitude> <longitude>Example:
GEOSET location1 Kolkata 22.5744 88.3629This will create a key called 'location1' and set it's name to "Kolkata". The coordinates would be stored as well. The data is stored on a 2-dimensional map. That means, you can add more values to the 'location1' key.
Upcoming commands:
GEOROUTE
GEODISTANCE
GEORADIUS
GEOELEVATIONMemify can act like a in memory database to compute time-series data.
If you're interested in contributing to the project, you can just fork the repo. If you're new to open source, check out how to contribute to a project