HeartSaVioR commented on a change in pull request #22138: [SPARK-25151][SS] Apply Apache Commons Pool to KafkaDataConsumer URL: https://github.com/apache/spark/pull/22138#discussion_r317774366
########## File path: external/kafka-0-10-sql/src/main/scala/org/apache/spark/sql/kafka010/InternalKafkaConsumerPool.scala ########## @@ -0,0 +1,226 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +package org.apache.spark.sql.kafka010 + +import java.{util => ju} +import java.util.concurrent.ConcurrentHashMap + +import org.apache.commons.pool2.{BaseKeyedPooledObjectFactory, PooledObject, SwallowedExceptionListener} +import org.apache.commons.pool2.impl.{DefaultEvictionPolicy, DefaultPooledObject, GenericKeyedObjectPool, GenericKeyedObjectPoolConfig} + +import org.apache.spark.SparkEnv +import org.apache.spark.internal.Logging +import org.apache.spark.sql.kafka010.InternalKafkaConsumerPool._ +import org.apache.spark.sql.kafka010.KafkaDataConsumer.CacheKey + +/** + * Provides object pool for [[InternalKafkaConsumer]] which is grouped by [[CacheKey]]. + * + * This class leverages [[GenericKeyedObjectPool]] internally, hence providing methods based on + * the class, and same contract applies: after using the borrowed object, you must either call + * returnObject() if the object is healthy to return to pool, or invalidateObject() if the object + * should be destroyed. + * + * The soft capacity of pool is determined by "spark.sql.kafkaConsumerCache.capacity" config value, + * and the pool will have reasonable default value if the value is not provided. + * (The instance will do its best effort to respect soft capacity but it can exceed when there's + * a borrowing request and there's neither free space nor idle object to clear.) + * + * This class guarantees that no caller will get pooled object once the object is borrowed and + * not yet returned, hence provide thread-safety usage of non-thread-safe [[InternalKafkaConsumer]] + * unless caller shares the object to multiple threads. + */ +private[kafka010] class InternalKafkaConsumerPool( + objectFactory: ObjectFactory, + poolConfig: PoolConfig) { + + // the class is intended to have only soft capacity + assert(poolConfig.getMaxTotal < 0) + + private lazy val pool = { Review comment: Just wanted to delay spending cost to when it's actually used, but now I think it's not a big deal as it's meant to be used fairly soon. Will remove lazy. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org
