Partitioning Sharding

Splitting data across multiple nodes/tables for scale — by what key, and what breaks when you shard.

Why we need this / what value this brings

Lets a dataset grow beyond what a single node can hold or serve efficiently.

Only once a single database node is a proven, measured bottleneck — premature sharding adds enormous complexity for no benefit at small scale.

Choose a shard key that matches your dominant query pattern (e.g. by region or provider_id) so most queries hit one shard.

Purely forward-looking for Localz at current scale — what would the natural shard key even be (region? provider_id?)?

Empty folder — drop notes, links, and findings here as you research.