rqlite

rqlite positions itself as a fault-tolerant relational database built on SQLite, aimed at people who want high availability without the usual operational overhead. Instead of adding a separate consensus layer or external services, it bundles replication into a single process, so a small cluster can be started by running the binary on a few machines.

The core workflow is straightforward: start a node with a data directory, join additional nodes to the cluster, and then issue SQL statements over HTTP using plain text or JSON. Because access is HTTP-based, there are no drivers to install, and client libraries are available for common languages. The system keeps nodes in sync and tolerates the failure of any single node without taking the database offline.

This makes rqlite most relevant to developers and small teams who appreciate SQLite’s simplicity but need redundancy, as well as to those building services where a lightweight, self-contained database cluster is preferable to a full-scale database server. It can also be useful for edge or embedded scenarios where a single binary and minimal dependencies matter.

When evaluating it, consider how your write volume and latency requirements fit a consensus-replicated SQLite setup, and test failover behavior in a staging cluster before relying on it in production. Installation options include a shell script, Homebrew, Docker, and Kubernetes, so it is easy to trial. Overall, rqlite is a pragmatic choice when simple clustering and H