TiDB is always an interesting project to see out in the wild!<p>However, anyone considering using it should be aware of its:<p><pre><code> - features: https://docs.pingcap.com/tidb/stable/basic-features
- limitations: https://docs.pingcap.com/tidb/stable/mysql-compatibility
</code></pre>
The former is a bit on the short side, while the latter is a somewhat long list. For example, TiDB doesn't support neither stored procedures, nor functions. It also doesn't have support for triggers and even for foreign key constraints, which would be an immediate dealbreaker to some.<p>That said, as long as you use the DB just as a medium for storing and querying data, without wanting to do too much in database processing, it could indeed be a good option! Now, why other databases don't make clustering so easy, is beyond me.<p>Personally, it feels like distributed systems are the future in one way or another, with clustering being available from get go - be it file systems like GlusterFS, data storage solutions like MongoDB, key value stores like Redis, or anything else. I have no illusions about distributed systems being easy to make, but setting them up easily (or even just running single node instances at first and later migrating to many instances) should be the norm.<p>Of course, i'm still obligated to drop this here: <a href="https://en.wikipedia.org/wiki/Fallacies_of_distributed_computing" rel="nofollow">https://en.wikipedia.org/wiki/Fallacies_of_distributed_compu...</a>