-
Notifications
You must be signed in to change notification settings - Fork 163
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[FEAT] Decouple pipeline building and running from new executor (#2522)
Co-authored-by: Colin Ho <[email protected]>
- Loading branch information
Showing
3 changed files
with
190 additions
and
101 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -1,5 +1,6 @@ | ||
| mod channel; | ||
| mod intermediate_ops; | ||
| mod pipeline; | ||
| mod run; | ||
| mod sinks; | ||
| mod sources; | ||
|
|
||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,185 @@ | ||
| use std::{collections::HashMap, sync::Arc}; | ||
|
|
||
| use daft_dsl::Expr; | ||
| use daft_micropartition::MicroPartition; | ||
| use daft_physical_plan::{ | ||
| Concat, Filter, InMemoryScan, Limit, LocalPhysicalPlan, PhysicalScan, Project, | ||
| UnGroupedAggregate, | ||
| }; | ||
| use daft_plan::populate_aggregation_stages; | ||
|
|
||
| use crate::{ | ||
| channel::MultiSender, | ||
| intermediate_ops::{ | ||
| aggregate::AggregateOperator, | ||
| filter::FilterOperator, | ||
| intermediate_op::{run_intermediate_op, IntermediateOperator}, | ||
| project::ProjectOperator, | ||
| }, | ||
| sinks::{ | ||
| aggregate::AggregateSink, | ||
| concat::ConcatSink, | ||
| limit::LimitSink, | ||
| sink::{run_double_input_sink, run_single_input_sink, DoubleInputSink, SingleInputSink}, | ||
| }, | ||
| sources::{ | ||
| in_memory::InMemorySource, | ||
| scan_task::ScanTaskSource, | ||
| source::{run_source, Source}, | ||
| }, | ||
| }; | ||
|
|
||
| pub enum PipelineNode { | ||
| Source { | ||
| source: Arc<dyn Source>, | ||
| }, | ||
| IntermediateOp { | ||
| intermediate_op: Box<dyn IntermediateOperator>, | ||
| child: Box<PipelineNode>, | ||
| }, | ||
| SingleInputSink { | ||
| sink: Box<dyn SingleInputSink>, | ||
| child: Box<PipelineNode>, | ||
| }, | ||
| DoubleInputSink { | ||
| sink: Box<dyn DoubleInputSink>, | ||
| left_child: Box<PipelineNode>, | ||
| right_child: Box<PipelineNode>, | ||
| }, | ||
| } | ||
|
|
||
| impl PipelineNode { | ||
| pub fn start(&self, sender: MultiSender) { | ||
| match self { | ||
| PipelineNode::Source { source } => { | ||
| run_source(source.clone(), sender); | ||
| } | ||
| PipelineNode::IntermediateOp { | ||
| intermediate_op, | ||
| child, | ||
| } => { | ||
| let sender = run_intermediate_op(intermediate_op.clone(), sender); | ||
| child.start(sender); | ||
| } | ||
| PipelineNode::SingleInputSink { sink, child } => { | ||
| let sender = run_single_input_sink(sink.clone(), sender); | ||
| child.start(sender); | ||
| } | ||
| PipelineNode::DoubleInputSink { | ||
| sink, | ||
| left_child, | ||
| right_child, | ||
| } => { | ||
| let (left_sender, right_sender) = run_double_input_sink(sink.clone(), sender); | ||
| left_child.start(left_sender); | ||
| right_child.start(right_sender); | ||
| } | ||
| } | ||
| } | ||
| } | ||
|
|
||
| pub fn physical_plan_to_pipeline( | ||
| physical_plan: &LocalPhysicalPlan, | ||
| psets: &HashMap<String, Vec<Arc<MicroPartition>>>, | ||
| ) -> PipelineNode { | ||
| match physical_plan { | ||
| LocalPhysicalPlan::PhysicalScan(PhysicalScan { scan_tasks, .. }) => { | ||
| let scan_task_source = ScanTaskSource::new(scan_tasks.clone()); | ||
| PipelineNode::Source { | ||
| source: Arc::new(scan_task_source), | ||
| } | ||
| } | ||
| LocalPhysicalPlan::InMemoryScan(InMemoryScan { info, .. }) => { | ||
| let partitions = psets.get(&info.cache_key).expect("Cache key not found"); | ||
| let in_memory_source = InMemorySource::new(partitions.clone()); | ||
| PipelineNode::Source { | ||
| source: Arc::new(in_memory_source), | ||
| } | ||
| } | ||
| LocalPhysicalPlan::Project(Project { | ||
| input, projection, .. | ||
| }) => { | ||
| let proj_op = ProjectOperator::new(projection.clone()); | ||
| let child_node = physical_plan_to_pipeline(input, psets); | ||
| PipelineNode::IntermediateOp { | ||
| intermediate_op: Box::new(proj_op), | ||
| child: Box::new(child_node), | ||
| } | ||
| } | ||
| LocalPhysicalPlan::Filter(Filter { | ||
| input, predicate, .. | ||
| }) => { | ||
| let filter_op = FilterOperator::new(predicate.clone()); | ||
| let child_node = physical_plan_to_pipeline(input, psets); | ||
| PipelineNode::IntermediateOp { | ||
| intermediate_op: Box::new(filter_op), | ||
| child: Box::new(child_node), | ||
| } | ||
| } | ||
| LocalPhysicalPlan::Limit(Limit { | ||
| input, num_rows, .. | ||
| }) => { | ||
| let sink = LimitSink::new(*num_rows as usize); | ||
| let child_node = physical_plan_to_pipeline(input, psets); | ||
| PipelineNode::SingleInputSink { | ||
| sink: Box::new(sink), | ||
| child: Box::new(child_node), | ||
| } | ||
| } | ||
| LocalPhysicalPlan::Concat(Concat { input, other, .. }) => { | ||
| let sink = ConcatSink::new(); | ||
| let left_child = physical_plan_to_pipeline(input, psets); | ||
| let right_child = physical_plan_to_pipeline(other, psets); | ||
| PipelineNode::DoubleInputSink { | ||
| sink: Box::new(sink), | ||
| left_child: Box::new(left_child), | ||
| right_child: Box::new(right_child), | ||
| } | ||
| } | ||
| LocalPhysicalPlan::UnGroupedAggregate(UnGroupedAggregate { | ||
| input, | ||
| aggregations, | ||
| schema, | ||
| .. | ||
| }) => { | ||
| let (first_stage_aggs, second_stage_aggs, final_exprs) = | ||
| populate_aggregation_stages(aggregations, schema, &[]); | ||
| let first_stage_agg_op = AggregateOperator::new( | ||
| first_stage_aggs | ||
| .values() | ||
| .cloned() | ||
| .map(|e| Arc::new(Expr::Agg(e.clone()))) | ||
| .collect(), | ||
| vec![], | ||
| ); | ||
| let second_stage_agg_sink = AggregateSink::new( | ||
| second_stage_aggs | ||
| .values() | ||
| .cloned() | ||
| .map(|e| Arc::new(Expr::Agg(e.clone()))) | ||
| .collect(), | ||
| vec![], | ||
| ); | ||
| let final_stage_project = ProjectOperator::new(final_exprs); | ||
|
|
||
| let child_node = physical_plan_to_pipeline(input, psets); | ||
| let intermediate_agg_op_node = PipelineNode::IntermediateOp { | ||
| intermediate_op: Box::new(first_stage_agg_op), | ||
| child: Box::new(child_node), | ||
| }; | ||
|
|
||
| let sink_node = PipelineNode::SingleInputSink { | ||
| sink: Box::new(second_stage_agg_sink), | ||
| child: Box::new(intermediate_agg_op_node), | ||
| }; | ||
|
|
||
| PipelineNode::IntermediateOp { | ||
| intermediate_op: Box::new(final_stage_project), | ||
| child: Box::new(sink_node), | ||
| } | ||
| } | ||
| _ => { | ||
| unimplemented!("Physical plan not supported: {}", physical_plan.name()); | ||
| } | ||
| } | ||
| } |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters