Module spill 

Disk spilling for queries that exceed available memory.

When a sort or aggregation grows too large for RAM, we spill partitions to disk and merge them back later. This lets queries complete even with limited memory - just slower.

Component	Purpose
SpillManager	Manages spill file lifecycle with automatic cleanup
SpillFile	Read/write individual spill files
ExternalSort	External merge sort for big ORDER BY
PartitionedState	Hash partitioning for spillable GROUP BY

Async variants (AsyncSpillManager, AsyncSpillFile) use tokio for non-blocking I/O when running in async contexts.

Structs

AsyncSpillFile

Async handle for a single spill file.

AsyncSpillFileReader

Async reader for a spill file.

AsyncSpillManager

Async manager for spill file lifecycle for out-of-core processing.

ExternalSort

External merge sort for out-of-core sorting.

PartitionedState

Partitioned accumulator state for spillable aggregation.

SortKey

Sort key specification.

SpillFile

Handle for a single spill file.

SpillFileReader

Reader for a spill file.

SpillManager

Manages spill file lifecycle for out-of-core processing.

Enums

NullOrder

Null handling in sort.

SortDirection

Sort direction.

Constants

DEFAULT_NUM_PARTITIONS

Default number of partitions for hash partitioning.

Functions

deserialize_row

Deserializes a row from bytes.

deserialize_value

Deserializes a Value from bytes.

serialize_row

Serializes a row (slice of Values) to bytes.

serialize_value

Serializes a Value to bytes.