shuffle/api/ShuffleExecutorComponents.java

0001 /*
0002  * Licensed to the Apache Software Foundation (ASF) under one or more
0003  * contributor license agreements.  See the NOTICE file distributed with
0004  * this work for additional information regarding copyright ownership.
0005  * The ASF licenses this file to You under the Apache License, Version 2.0
0006  * (the "License"); you may not use this file except in compliance with
0007  * the License.  You may obtain a copy of the License at
0008  *
0009  *    http://www.apache.org/licenses/LICENSE-2.0
0010  *
0011  * Unless required by applicable law or agreed to in writing, software
0012  * distributed under the License is distributed on an "AS IS" BASIS,
0013  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
0014  * See the License for the specific language governing permissions and
0015  * limitations under the License.
0016  */
0017
0018 package org.apache.spark.shuffle.api;
0019
0020 import java.io.IOException;
0021 import java.util.Map;
0022 import java.util.Optional;
0023
0024 import org.apache.spark.annotation.Private;
0025
0026 /**
0027  * :: Private ::
0028  * An interface for building shuffle support for Executors.
0029  *
0030  * @since 3.0.0
0031  */
0032 @Private
0033 public interface ShuffleExecutorComponents {
0034
0035   /**
0036    * Called once per executor to bootstrap this module with state that is specific to
0037    * that executor, specifically the application ID and executor ID.
0038    *
0039    * @param appId The Spark application id
0040    * @param execId The unique identifier of the executor being initialized
0041    * @param extraConfigs Extra configs that were returned by
0042    *                     {@link ShuffleDriverComponents#initializeApplication()}
0043    */
0044   void initializeExecutor(String appId, String execId, Map<String, String> extraConfigs);
0045
0046   /**
0047    * Called once per map task to create a writer that will be responsible for persisting all the
0048    * partitioned bytes written by that map task.
0049    *
0050    * @param shuffleId Unique identifier for the shuffle the map task is a part of
0051    * @param mapTaskId An ID of the map task. The ID is unique within this Spark application.
0052    * @param numPartitions The number of partitions that will be written by the map task. Some of
0053    *                      these partitions may be empty.
0054    */
0055   ShuffleMapOutputWriter createMapOutputWriter(
0056       int shuffleId,
0057       long mapTaskId,
0058       int numPartitions) throws IOException;
0059
0060   /**
0061    * An optional extension for creating a map output writer that can optimize the transfer of a
0062    * single partition file, as the entire result of a map task, to the backing store.
0063    * <p>
0064    * Most implementations should return the default {@link Optional#empty()} to indicate that
0065    * they do not support this optimization. This primarily is for backwards-compatibility in
0066    * preserving an optimization in the local disk shuffle storage implementation.
0067    *
0068    * @param shuffleId Unique identifier for the shuffle the map task is a part of
0069    * @param mapId An ID of the map task. The ID is unique within this Spark application.
0070    */
0071   default Optional<SingleSpillShuffleMapOutputWriter> createSingleFileMapOutputWriter(
0072       int shuffleId,
0073       long mapId) throws IOException {
0074     return Optional.empty();
0075   }
0076 }