Skip to content

Commit

Permalink
organize ExternalSorter fields (#13447)
Browse files Browse the repository at this point in the history
  • Loading branch information
2010YOUY01 authored Nov 16, 2024
1 parent 06db9ed commit 73507c3
Showing 1 changed file with 29 additions and 14 deletions.
43 changes: 29 additions & 14 deletions datafusion/physical-plan/src/sorts/sort.rs
Original file line number Diff line number Diff line change
Expand Up @@ -203,39 +203,54 @@ impl ExternalSorterMetrics {
/// in_mem_batches
/// ```
struct ExternalSorter {
/// schema of the output (and the input)
// ========================================================================
// PROPERTIES:
// Fields that define the sorter's configuration and remain constant
// ========================================================================
/// Schema of the output (and the input)
schema: SchemaRef,
/// Sort expressions
expr: Arc<[PhysicalSortExpr]>,
/// If Some, the maximum number of output rows that will be produced
fetch: Option<usize>,
/// The target number of rows for output batches
batch_size: usize,
/// If the in size of buffered memory batches is below this size,
/// the data will be concatenated and sorted in place rather than
/// sort/merged.
sort_in_place_threshold_bytes: usize,

// ========================================================================
// STATE BUFFERS:
// Fields that hold intermediate data during sorting
// ========================================================================
/// Potentially unsorted in memory buffer
in_mem_batches: Vec<RecordBatch>,
/// if `Self::in_mem_batches` are sorted
in_mem_batches_sorted: bool,

/// If data has previously been spilled, the locations of the
/// spill files (in Arrow IPC format)
spills: Vec<RefCountedTempFile>,
/// Sort expressions
expr: Arc<[PhysicalSortExpr]>,

// ========================================================================
// EXECUTION RESOURCES:
// Fields related to managing execution resources and monitoring performance.
// ========================================================================
/// Runtime metrics
metrics: ExternalSorterMetrics,
/// If Some, the maximum number of output rows that will be
/// produced.
fetch: Option<usize>,
/// A handle to the runtime to get spill files
runtime: Arc<RuntimeEnv>,
/// Reservation for in_mem_batches
reservation: MemoryReservation,

/// Reservation for the merging of in-memory batches. If the sort
/// might spill, `sort_spill_reservation_bytes` will be
/// pre-reserved to ensure there is some space for this sort/merge.
merge_reservation: MemoryReservation,
/// A handle to the runtime to get spill files
runtime: Arc<RuntimeEnv>,
/// The target number of rows for output batches
batch_size: usize,
/// How much memory to reserve for performing in-memory sort/merges
/// prior to spilling.
sort_spill_reservation_bytes: usize,
/// If the in size of buffered memory batches is below this size,
/// the data will be concatenated and sorted in place rather than
/// sort/merged.
sort_in_place_threshold_bytes: usize,
}

impl ExternalSorter {
Expand Down

0 comments on commit 73507c3

Please sign in to comment.