Package org.apache.ignite.example.compute

Class ComputeMapReduceExample.PhraseWordLengthCountMapReduceTask

java.lang.Object

org.apache.ignite.example.compute.ComputeMapReduceExample.PhraseWordLengthCountMapReduceTask

All Implemented Interfaces:

MapReduceTask<String,String,Integer,Integer>

Enclosing class:

ComputeMapReduceExample

public static class ComputeMapReduceExample.PhraseWordLengthCountMapReduceTask
extends Object
implements MapReduceTask<String,String,Integer,Integer>

MapReduce task that splits the input phrase into words and sends them to different nodes for the processing.

Constructor Summary

Constructors

Constructor	Description
PhraseWordLengthCountMapReduceTask()

Method Summary

Modifier and Type	Method	Description
CompletableFuture<Integer>	reduceAsync(TaskExecutionContext taskContext, Map<UUID,Integer> results)	This is a finishing step in the task execution.
CompletableFuture<List<MapReduceJob<String,Integer>>>	splitAsync(TaskExecutionContext taskContext, String input)	This method should return a list of compute job execution parameters which will be used to submit compute jobs.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.apache.ignite.compute.task.MapReduceTask

reduceJobResultMarshaller, splitJobInputMarshaller

Constructor Details

PhraseWordLengthCountMapReduceTask

public PhraseWordLengthCountMapReduceTask()

Method Details

splitAsync

public CompletableFuture<List<MapReduceJob<String,Integer>>> splitAsync(TaskExecutionContext taskContext,
 String input)

This method should return a list of compute job execution parameters which will be used to submit compute jobs.

Specified by:

splitAsync in interface MapReduceTask<String,String,Integer,Integer>

Parameters:

taskContext - Task execution context.

input - Map reduce task (I)nput.

Returns:

A future with the list of compute job execution parameters.

reduceAsync

public CompletableFuture<Integer> reduceAsync(TaskExecutionContext taskContext,
 Map<UUID,Integer> results)

This is a finishing step in the task execution. This method will be called with the map from identifiers of compute jobs submitted as a result of the MapReduceTask.splitAsync(TaskExecutionContext, Object) method call to the results of the execution of the corresponding job. The resulting map has the same iteration order as the result of the split method. The return value of this method will be returned as a result of this task.

Specified by:

reduceAsync in interface MapReduceTask<String,String,Integer,Integer>

Parameters:

taskContext - Task execution context.

results - Map from compute job ids to their results.

Returns:

Final task result future.