MergeTree from First Principles
MechanicsBuild a correct mental model of how MergeTree actually stores and finds data, from first principles.
What this module covers
- MergeTree table engine mechanics
- Parts (immutable, created per insert)
- Partitions (PARTITION BY)
- Background merges and merge selection
- Sparse primary index and granules (index_granularity)
- ORDER BY as the (mostly) primary key
- Data skipping indexes (minmax, set, bloom filter)