yew1eb commented on code in PR #3793: URL: https://github.com/apache/celeborn/pull/3793#discussion_r3803796459
########## client-spark/spark-3-ui/src/main/scala/org/apache/spark/shuffle/celeborn/CelebornListener.scala: ########## @@ -0,0 +1,101 @@ +/* + * 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.shuffle.celeborn + +import java.util.concurrent.atomic.AtomicLong + +import org.apache.spark.SparkConf +import org.apache.spark.internal.Logging +import org.apache.spark.scheduler._ +import org.apache.spark.util.kvstore.KVStore + +/** + * Collects Celeborn shuffle metrics into the Spark KVStore for live UI and + * HistoryServer replay. + */ +private[celeborn] class CelebornListener( + val kvstore: KVStore, + val conf: SparkConf) + extends SparkListener with Logging { + + private val totalWriteBytes = new AtomicLong(0L) + private val totalWriteTimeMs = new AtomicLong(0L) + private val totalReadBytes = new AtomicLong(0L) + private val totalFetchWaitTimeMs = new AtomicLong(0L) + private val totalTaskDurationMs = new AtomicLong(0L) + + private val lastUpdateTimestamp = new AtomicLong(-1L) + private val updateIntervalMillis = 5000L + + def register(sc: org.apache.spark.SparkContext): Unit = { + sc.addSparkListener(this) + logInfo("CelebornListener registered successfully") + } + + override def onTaskEnd(taskEnd: SparkListenerTaskEnd): Unit = { + Option(taskEnd.taskMetrics).foreach { metrics => + totalWriteBytes.addAndGet(metrics.shuffleWriteMetrics.bytesWritten) + // writeTime is in nanoseconds; normalize to ms. + totalWriteTimeMs.addAndGet(metrics.shuffleWriteMetrics.writeTime / 1000000L) + totalReadBytes.addAndGet(metrics.shuffleReadMetrics.totalBytesRead) + totalFetchWaitTimeMs.addAndGet(metrics.shuffleReadMetrics.fetchWaitTime) + totalTaskDurationMs.addAndGet(taskEnd.taskInfo.duration) + } + mayUpdate() + } + + override def onEnvironmentUpdate(environmentUpdate: SparkListenerEnvironmentUpdate): Unit = { + val celebornProps = environmentUpdate.environmentDetails + .getOrElse("Spark Properties", Seq.empty) + .filter { case (k, _) => k.startsWith("spark.celeborn.") } + .sortBy(_._1) + if (celebornProps.nonEmpty) { + kvstore.write(new CelebornPropertiesUIData(celebornProps.toList)) + } + } + + override def onApplicationEnd(applicationEnd: SparkListenerApplicationEnd): Unit = { + mayUpdate(force = true) + logInfo("CelebornListener: application ended, final flush completed") + } + + private def mayUpdate(force: Boolean = false): Unit = { + val now = System.currentTimeMillis() + val last = lastUpdateTimestamp.get() + if (!force && (last != -1L && (now - last) < updateIntervalMillis)) { + return + } + if (lastUpdateTimestamp.compareAndSet(last, now) || force) { + flushAggregations() + } Review Comment: Impact is negligible in practice: `force=true` is only reached from `onApplicationEnd`, which fires once at app end with no concurrent `onTaskEnd`. No extra flushes follow it. Leaving as-is. -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
