You probably had the problem before. You created your backend with a great database but found that you can’t shard it. I’m going to share how I met these problems and solved them with a database built for the purpose.
# The Beginning
I wanted a database that was quite small and fast, but supported clustering. I wanted to use LevelDB, but it can’t support clustering. After some time, I discovered that it was possible to create a IPC socket and shard the database across multiple processes.
I became so excited that I started to work on a small project called ReziDB. It quickly evolved to include features such as clustering, querying, and at the top, speed.
# The Release
Recently, ReziDB walked into the world as a GitHub repository and a NPM package. It fills the gap between huge, remote databases and the small JSON-based databases.
# The Result
ReziDB had outstanding results. It maxxed out at around 70,000 ops/s (get) and 56,000 ops/s (set). In cluster mode, it ran around ~6,000 ops/s slower, but still held the amazing speed.
# The future
ReziDB is still a live project and is due for quite a few features. A few of the top features would be a UDP/HTTP connection, sharding across machines, a real dashboard and security, and more.
