Struct NetworkBroadcastExec 

pub struct NetworkBroadcastExec { /* private fields */ }

Network boundary for broadcasting data to all consumer tasks.

This operator works with BroadcastExec which scales up partitions so each consumer task fetches a unique set of partition numbers. Each partition request is sent to all stage tasks because each task’s leaf node is specialized to serve a different slice of the data for the same logical partition number.

Here are some examples of how NetworkBroadcastExec distributes data:

1 to many

┌────────────────────────┐                        ┌────────────────────────┐           ■
│  NetworkBroadcastExec  │                        │  NetworkBroadcastExec  │           │
│        (task 1)        │           ...          │        (task M)        │           │
│                        │                        │                        │        Stage N
│    Populates Caches    │                        │    Populates Caches    │           │
└────────┬─┬┬─┬┬─┬───────┘                        └────────┬─┬┬─┬┬─┬───────┘           │
         │0││1││2│                                         │0││1││2│                   │
         └▲┘└▲┘└▲┘                                         └▲┘└▲┘└▲┘                   ■
          │  │  │                                           │  │  │
          │  │  │                                           │  │  │
          │  │  │                                           │  │  │
          │  │  └─────────────┐          ┌──────────────────┘  │  │
          │  └─────────────┐  │          │     ┌───────────────┘  │
          └─────────────┐  │  │          │     │    ┌─────────────┘
                        │  │  │          │     │    │
                       ┌┴┐┌┴┐┌┴┐ ... ┌───┴┐┌───┴┐┌──┴─┐
                       │1││2││3│     │NM-3││NM-2││NM-1│                                ■
                      ┌┴─┴┴─┴┴─┴─────┴────┴┴────┴┴────┴─┐                              │
                      │          BroadcastExec          │                              │
                      │        ┌───────────────┐        │                          Stage N-1
                      │        │  Batch Cache  │        │                              │
                      │        │  ┌─┐ ┌─┐ ┌─┐  │        │                              │
                      │        │  │0│ │1│ │2│  │        │                              │
                      │        │  └─┘ └─┘ └─┘  │        │                              │
                      │        └───────────────┘        │                              │
                      └───────────┬─┬─┬─┬─┬─┬───────────┘                              │
                                  │0│ │1│ │2│                                          │
                                  └▲┘ └▲┘ └▲┘                                          ■
                                   │   │   │
                                   │   │   │
                                   │   │   │
                                  ┌┴┐ ┌┴┐ ┌┴┐                                          ■
                                  │0│ │1│ │2│                                          │
                           ┌──────┴─┴─┴─┴─┴─┴──────┐                               Stage N-2
                           │Arc<dyn ExecutionPlan> │                                   │
                           │       (task 1)        │                                   │
                           └───────────────────────┘                                   ■

Many to many

   ┌────────────────────────┐                        ┌────────────────────────┐          ■
   │  NetworkBroadcastExec  │                        │  NetworkBroadcastExec  │          │
   │        (task 1)        │                        │        (task M)        │          │
   │                        │           ...          │                        │       Stage N
   │    Populates Caches    │                        │       Cache Hits       │          │
   └────────┬─┬┬─┬┬─┬───────┘                        └────────┬─┬┬─┬┬─┬───────┘          │
            │0││1││2│                                         │0││1││2│                  │
            └▲┘└▲┘└▲┘                                         └▲┘└▲┘└▲┘                  ■
             │  │  │                                           │  │  │
  ┌──────────┴──┼──┼────────────────────────────────┐          │  │  │
  │  ┌──────────┴──┼────────────────────────────────┼──┐       │  │  │
  │  │  ┌──────────┴────────────────────────────────┼──┼──┐    │  │  │
  │  │  │                                           │  │  │    │  │  │
  │  │  │         ┌─────────────────────────────────┼──┼──┼────┴──┼─┐│
  │  │  │         │     ┌───────────────────────────┼──┼──┼───────┴─┼┼─────┐
  │  │  │         │     │     ┌─────────────────────┼──┼──┼─────────┼┴─────┼────┐
  │  │  │         │     │     │                     │  │  │         │      │    │
 ┌┴┐┌┴┐┌┴┐ ... ┌──┴─┐┌──┴─┐┌──┴─┐                  ┌┴┐┌┴┐┌┴┐ ... ┌──┴─┐┌───┴┐┌──┴─┐      ■
 │0││1││2│     │3M-3││3M-2││3M-1│                  │0││1││2│     │3M-3││3M-2││3M-1│      │
┌┴─┴┴─┴┴─┴─────┴────┴┴────┴┴────┴┐                ┌┴─┴┴─┴┴─┴─────┴────┴┴────┴┴────┴┐     │
│         BroadcastExec          │                │         BroadcastExec          │     │
│        ┌───────────────┐       │                │        ┌───────────────┐       │     │
│        │  Batch Cache  │       │                │        │  Batch Cache  │       │     │
│        │  ┌─┐ ┌─┐ ┌─┐  │       │      ...       │        │  ┌─┐ ┌─┐ ┌─┐  │       │ Stage N-1
│        │  │0│ │1│ │2│  │       │                │        │  │0│ │1│ │2│  │       │     │
│        │  └─┘ └─┘ └─┘  │       │                │        │  └─┘ └─┘ └─┘  │       │     │
│        └───────────────┘       │                │        └───────────────┘       │     │
└───────────┬─┬─┬─┬─┬─┬──────────┘                └───────────┬─┬─┬─┬─┬─┬──────────┘     │
            │0│ │1│ │2│                                       │0│ │1│ │2│                │
            └▲┘ └▲┘ └▲┘                                       └▲┘ └▲┘ └▲┘                ■
             │   │   │                                         │   │   │
             │   │   │                                         │   │   │
             │   │   │                                         │   │   │
            ┌┴┐ ┌┴┐ ┌┴┐                                       ┌┴┐ ┌┴┐ ┌┴┐                ■
            │0│ │1│ │2│                                       │0│ │1│ │2│                │
     ┌──────┴─┴─┴─┴─┴─┴──────┐                         ┌──────┴─┴─┴─┴─┴─┴──────┐     Stage N-2
     │Arc<dyn ExecutionPlan> │          ...            │Arc<dyn ExecutionPlan> │         │
     │       (task 1)        │                         │       (task N)        │         │
     └───────────────────────┘                         └───────────────────────┘         ■

Notice in this diagram that each NetworkBroadcastExec sends a request to fetch data from each BroadcastExec in the stage below per partition. This is because each BroadcastExec has its own cache which contains partial results for the partition. It is the NetworkBroadcastExec’s job to merge these partial partitions to then broadcast complete data to the consumers.

Implementations

impl NetworkBroadcastExec

pub fn try_new( input: Arc<dyn ExecutionPlan>, producer_tasks: usize, ) -> Result<Self>

Creates a new NetworkBroadcastExec fed by the provided BroadcastExec. The input plan will be executed in a remote worker in producer_tasks number of tasks.

Trait Implementations

impl Clone for NetworkBroadcastExec

fn clone(&self) -> NetworkBroadcastExec

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for NetworkBroadcastExec

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl DisplayAs for NetworkBroadcastExec

fn fmt_as(&self, _t: DisplayFormatType, f: &mut Formatter<'_>) -> Result

Format according to DisplayFormatType, used when verbose representation looks different from the default one

impl ExecutionPlan for NetworkBroadcastExec

fn name(&self) -> &str

Short name for the ExecutionPlan, such as ‘DataSourceExec’.

fn properties(&self) -> &Arc<PlanProperties>

Return properties of the output of the ExecutionPlan, such as output ordering(s), partitioning information etc.

fn children(&self) -> Vec<&Arc<dyn ExecutionPlan>>

Get a list of children ExecutionPlans that act as inputs to this plan. The returned list will be empty for leaf nodes such as scans, will contain a single value for unary nodes, or two values for binary nodes (such as joins).

fn with_new_children( self: Arc<Self>, children: Vec<Arc<dyn ExecutionPlan>>, ) -> Result<Arc<dyn ExecutionPlan>, DataFusionError>

Returns a new ExecutionPlan where all existing children were replaced by the children, in order

fn execute( &self, partition: usize, context: Arc<TaskContext>, ) -> Result<SendableRecordBatchStream, DataFusionError>

Begin execution of partition, returning a Stream of RecordBatches.

fn metrics(&self) -> Option<MetricsSet>

Return a snapshot of the set of Metrics for this ExecutionPlan. If no Metrics are available, return None.

fn static_name() -> &'static str

where Self: Sized,

Short name for the ExecutionPlan, such as ‘DataSourceExec’. Like name but can be called without an instance.

fn downcast_delegate(&self) -> Option<&(dyn ExecutionPlan + 'static)>

Returns the plan that provides this plan’s public ExecutionPlan downcast identity.

fn schema(&self) -> Arc<Schema>

Get the schema for this execution plan

fn check_invariants(&self, check: InvariantLevel) -> Result<(), DataFusionError>

Returns an error if this individual node does not conform to its invariants. These invariants are typically only checked in debug mode.

fn required_input_distribution(&self) -> Vec<Distribution>

Specifies the data distribution requirements for all the children for this ExecutionPlan, By default it’s [Distribution::UnspecifiedDistribution] for each child,

fn required_input_ordering(&self) -> Vec<Option<OrderingRequirements>>

Specifies the ordering required for all of the children of this ExecutionPlan.

fn maintains_input_order(&self) -> Vec<bool>

Returns false if this ExecutionPlan’s implementation may reorder rows within or between partitions.

fn benefits_from_input_partitioning(&self) -> Vec<bool>

Specifies whether the ExecutionPlan benefits from increased parallelization at its input for each child.

fn reset_state( self: Arc<Self>, ) -> Result<Arc<dyn ExecutionPlan>, DataFusionError>

Reset any internal state within this ExecutionPlan.

fn repartitioned( &self, _target_partitions: usize, _config: &ConfigOptions, ) -> Result<Option<Arc<dyn ExecutionPlan>>, DataFusionError>

If supported, attempt to increase the partitioning of this ExecutionPlan to produce target_partitions partitions.

fn partition_statistics( &self, partition: Option<usize>, ) -> Result<Arc<Statistics>, DataFusionError>

Returns statistics for a specific partition of this ExecutionPlan node. If statistics are not available, should return Statistics::new_unknown (the default), not an error. If partition is None, it returns statistics for the entire plan.

fn supports_limit_pushdown(&self) -> bool

Returns true if a limit can be safely pushed down through this ExecutionPlan node.

fn with_fetch(&self, _limit: Option<usize>) -> Option<Arc<dyn ExecutionPlan>>

Returns a fetching variant of this ExecutionPlan node, if it supports fetch limits. Returns None otherwise.

fn fetch(&self) -> Option<usize>

Gets the fetch count for the operator, None means there is no fetch.

fn cardinality_effect(&self) -> CardinalityEffect

Gets the effect on cardinality, if known

fn try_swapping_with_projection( &self, _projection: &ProjectionExec, ) -> Result<Option<Arc<dyn ExecutionPlan>>, DataFusionError>

Attempts to push down the given projection into the input of this ExecutionPlan.

fn gather_filters_for_pushdown( &self, _phase: FilterPushdownPhase, parent_filters: Vec<Arc<dyn PhysicalExpr>>, _config: &ConfigOptions, ) -> Result<FilterDescription, DataFusionError>

Collect filters that this node can push down to its children. Filters that are being pushed down from parents are passed in, and the node may generate additional filters to push down. For example, given the plan FilterExec -> HashJoinExec -> DataSourceExec, what will happen is that we recurse down the plan calling ExecutionPlan::gather_filters_for_pushdown:

fn handle_child_pushdown_result( &self, _phase: FilterPushdownPhase, child_pushdown_result: ChildPushdownResult, _config: &ConfigOptions, ) -> Result<FilterPushdownPropagation<Arc<dyn ExecutionPlan>>, DataFusionError>

Handle the result of a child pushdown.

fn with_new_state( &self, _state: Arc<dyn Any + Send + Sync>, ) -> Option<Arc<dyn ExecutionPlan>>

Injects arbitrary run-time state into this execution plan, returning a new plan instance that incorporates that state if it is relevant to the concrete node implementation.

fn try_pushdown_sort( &self, _order: &[PhysicalSortExpr], ) -> Result<SortOrderPushdownResult<Arc<dyn ExecutionPlan>>, DataFusionError>

Try to push down sort ordering requirements to this node.

fn with_preserve_order( &self, _preserve_order: bool, ) -> Option<Arc<dyn ExecutionPlan>>

Returns a variant of this ExecutionPlan that is aware of order-sensitivity.

impl NetworkBoundary for NetworkBroadcastExec

fn with_input_stage(&self, input_stage: Stage) -> Result<Arc<dyn ExecutionPlan>>

Called when a Stage is correctly formed. The NetworkBoundary can use this information to perform any internal transformations necessary for distributed execution.

fn input_stage(&self) -> &Stage

Returns the assigned input Stage, if any.

fn producer_head(&self, consumer_task_count: usize) -> ProducerHead

Defines what head node should the producer stage feeding this NetworkBoundary implementation have. This information is used during planning an executing for ensuring the head of a stage has the appropriate shape for consumption.