Struct DistributedPhysicalOptimizerRule 

pub struct DistributedPhysicalOptimizerRule;

Physical optimizer rule that inspects the plan, places the appropriate network boundaries, and breaks it down into stages that can be executed in a distributed manner.

The rule has three steps:

1. Annotate the plan with [annotate_plan]: adds some annotations to each node about how many distributed tasks should be used in the stage containing them, and whether they need a network boundary below or not. For more information about this step, read [annotate_plan] docs.

2. Based on the [AnnotatedPlan] returned by [annotate_plan], place all the appropriate network boundaries (NetworkShuffleExec and NetworkCoalesceExec) with the task count assignation that the annotations required. After this, the plan is already a distributed executable plan.

3. Place the [CoalesceBatchesExec] in the appropriate places (just below network boundaries), so that we send fewer and bigger record batches over the wire instead of a lot of small ones.

Trait Implementations

impl Debug for DistributedPhysicalOptimizerRule

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

Formats the value using the given formatter.

impl Default for DistributedPhysicalOptimizerRule

fn default() -> DistributedPhysicalOptimizerRule

Returns the “default value” for a type.

impl PhysicalOptimizerRule for DistributedPhysicalOptimizerRule

fn optimize( &self, original: Arc<dyn ExecutionPlan>, cfg: &ConfigOptions, ) -> Result<Arc<dyn ExecutionPlan>>

Rewrite plan to an optimized form

fn name(&self) -> &str

A human readable name for this optimizer rule

fn schema_check(&self) -> bool

A flag to indicate whether the physical planner should validate that the rule will not change the schema of the plan after the rewriting. Some of the optimization rules might change the nullable properties of the schema and should disable the schema check.