Definition
Range partitioning divides data by ordered key ranges. Each shard holds a contiguous range of key values.
Core concept
Good for range queries: all values in a range reside together. For instance, shard1 holds keys 0-1000, shard2 1001-2000, etc.
Use cases
Partitioning logs by date (Jan 2026 in one partition).
Trade-offs
Can lead to hot shards if data skews. Splitting ranges as data grows requires resharding.
References
Sharding tutorials.