A fast-to-sync/search and space-optimized replication algorithm written in rust, The Nun-db data replication model – Mateus Freira – Lead Software Engineer at @RadicalImaging and MSc Computer Science from PucMinas
I have been working on Nun-db as a side project for over two years. Finally, in June 2020, I got it running as the unique real-time database for my main application. In the post, I shared that milestone. Since then, a couple of other applications have been starting using Nun-db, e.g., a chatbot, a real-time visitor for my blog, a house price scraper, a tic tac toe multiplayer, etc.
Most of them are small applications but they were good enough to test the stability and the main concepts of Nun-db as a viable option. So it was now time for the next step: make it ready to scale and safe to use in more prominent use-cases; that means not trusting in a single VM to keep it running, that means making it distributed, that means lots of new challenges and learnings. So this is the first post of a series of how I am making that possible.
Replicate data is essential for several reasons. Disaster recovery and high availability are probably the most important ones for databases. In addition, data replication is crucial for Nun-db to help drive the adoption of applications with more critical loads and serve them as their primary database.
