HeartSaVioR commented on code in PR #43393:
URL: https://github.com/apache/spark/pull/43393#discussion_r1362048469
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/state/OperatorStateMetadata.scala:
##########
Review Comment:
General comment: we'd like to have a class doc for each trait/class.
##########
sql/core/src/main/scala/org/apache/spark/sql/execution/streaming/IncrementalExecution.scala:
##########
@@ -177,6 +183,20 @@ class IncrementalExecution(
}
}
+ object WriteStatefulOperatorMetadataRule extends SparkPlanPartialRule {
+ override val rule: PartialFunction[SparkPlan, SparkPlan] = {
+ case stateStoreWriter: StateStoreWriter =>
+ if (isFirstBatch) {
+ val metadata = stateStoreWriter.operatorStateMetadata()
+ val metadataWriter = new OperatorStateMetadataWriter(new Path(
+ checkpointLocation,
stateStoreWriter.getStateInfo.operatorId.toString), hadoopConf)
+ metadataWriter.write(metadata)
+ }
+ stateStoreWriter
+ case plan: SparkPlan => plan
Review Comment:
nit: Does this require all cases to be matched? I guess it should be OK to
construct a "partial" match since we are creating PartialFunction. This line
may not be needed.
##########
sql/core/src/test/scala/org/apache/spark/sql/execution/streaming/state/OperatorStateMetadataSuite.scala:
##########
@@ -0,0 +1,158 @@
+/*
+ * 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.execution.streaming.state
+
+import org.apache.hadoop.fs.Path
+
+import org.apache.spark.sql.Column
+import org.apache.spark.sql.execution.streaming.MemoryStream
+import org.apache.spark.sql.functions.{count, session_window}
+import org.apache.spark.sql.streaming.{OutputMode, StreamTest}
+import org.apache.spark.sql.streaming.OutputMode.Complete
+import org.apache.spark.sql.test.SharedSparkSession
+import org.apache.spark.util.Utils
+
+class OperatorStateMetadataSuite extends StreamTest with SharedSparkSession {
+ import testImplicits._
+
+ private lazy val hadoopConf = spark.sessionState.newHadoopConf()
+
+ private def numShufflePartitions =
spark.sessionState.conf.numShufflePartitions
+
+ private def sameOperatorStateMetadata(
+ operatorMetadata1: OperatorStateMetadataV1,
+ operatorMetadata2: OperatorStateMetadataV1): Boolean = {
+ operatorMetadata1.operatorInfo == operatorMetadata2.operatorInfo &&
+
operatorMetadata1.stateStoreInfo.sameElements(operatorMetadata2.stateStoreInfo)
+ }
+
+ test("Serialize and deserialize stateful operator metadata") {
+ val stateDir = Utils.createTempDir()
+ val statePath = new Path(stateDir.toString)
+ val stateStoreInfo = (1 to 4).map(i => StateStoreMetadataV1(s"store$i", 1,
200))
+ val operatorInfo = OperatorInfoV1(1, "Join")
+ val operatorMetadata = OperatorStateMetadataV1(operatorInfo,
stateStoreInfo.toArray)
+ new OperatorStateMetadataWriter(statePath,
hadoopConf).write(operatorMetadata)
+ val operatorMetadata1 = new OperatorStateMetadataReader(statePath,
hadoopConf).read()
+ .asInstanceOf[OperatorStateMetadataV1]
+ assert(sameOperatorStateMetadata(operatorMetadata, operatorMetadata1))
+ }
+
+ test("Stateful operator metadata for streaming aggregation") {
+ val inputData = MemoryStream[Int]
+ val checkpointDir = Utils.createTempDir()
+ val aggregated =
+ inputData.toDF()
+ .groupBy($"value")
+ .agg(count("*"))
+ .as[(Int, Long)]
+
+ testStream(aggregated, Complete)(
+ StartStream(checkpointLocation = checkpointDir.toString),
+ AddData(inputData, 3),
+ CheckLastBatch((3, 1)),
+ StopStream
+ )
+
+ val statePath = new Path(checkpointDir.getCanonicalPath, "state/0")
+ val operatorMetadata = new OperatorStateMetadataReader(statePath,
hadoopConf).read()
+ .asInstanceOf[OperatorStateMetadataV1]
+ val expectedMetadata = OperatorStateMetadataV1(OperatorInfoV1(0,
"stateStoreSave"),
+ Array(StateStoreMetadataV1("default", 0, numShufflePartitions)))
+ assert(sameOperatorStateMetadata(operatorMetadata, expectedMetadata))
+ }
+
+ test("Stateful operator metadata for streaming join") {
+ val input1 = MemoryStream[Int]
+ val input2 = MemoryStream[Int]
+
+ val df1 = input1.toDF.select($"value" as "key", ($"value" * 2) as
"leftValue")
+ val df2 = input2.toDF.select($"value" as "key", ($"value" * 3) as
"rightValue")
+ val joined = df1.join(df2, "key")
+
+ val checkpointDir = Utils.createTempDir()
+ testStream(joined)(
+ StartStream(checkpointLocation = checkpointDir.getCanonicalPath),
+ AddData(input1, 1),
+ CheckAnswer(),
+ AddData(input2, 1, 10), // 1 arrived on input1 first, then input2,
should join
+ CheckNewAnswer((1, 2, 3)),
+ StopStream
+ )
+
+ val statePath = new Path(checkpointDir.toString, "state/0")
+ val operatorMetadata = new OperatorStateMetadataReader(statePath,
hadoopConf).read()
+ .asInstanceOf[OperatorStateMetadataV1]
+
+ val expectedStateStoreInfo = Array(
+ StateStoreMetadataV1("left-keyToNumValues", 0, numShufflePartitions),
+ StateStoreMetadataV1("left-keyWithIndexToValue", 0,
numShufflePartitions),
+ StateStoreMetadataV1("right-keyToNumValues", 0, numShufflePartitions),
+ StateStoreMetadataV1("right-keyWithIndexToValue", 0,
numShufflePartitions))
+
+ val expectedMetadata = OperatorStateMetadataV1(
+ OperatorInfoV1(0, "symmetricHashJoin"), expectedStateStoreInfo)
+ assert(sameOperatorStateMetadata(operatorMetadata, expectedMetadata))
+ }
+
+ test("Stateful operator metadata for streaming session window") {
+ val input = MemoryStream[(String, Long)]
+ val sessionWindow: Column = session_window($"eventTime", "10 seconds")
+
+ val checkpointDir = Utils.createTempDir()
+
+ val events = input.toDF()
+ .select($"_1".as("value"), $"_2".as("timestamp"))
+ .withColumn("eventTime", $"timestamp".cast("timestamp"))
+ .withWatermark("eventTime", "30 seconds")
+ .selectExpr("explode(split(value, ' ')) AS sessionId", "eventTime")
+
+ val streamingDf = events
+ .groupBy(sessionWindow as Symbol("session"), $"sessionId")
+ .agg(count("*").as("numEvents"))
+ .selectExpr("sessionId", "CAST(session.start AS LONG)",
"CAST(session.end AS LONG)",
+ "CAST(session.end AS LONG) - CAST(session.start AS LONG) AS
durationMs",
+ "numEvents")
+
+ testStream(streamingDf, OutputMode.Complete())(
+ StartStream(checkpointLocation = checkpointDir.toString),
+ AddData(input,
+ ("hello world spark streaming", 40L),
+ ("world hello structured streaming", 41L)
+ ),
+ CheckNewAnswer(
+ ("hello", 40, 51, 11, 2),
+ ("world", 40, 51, 11, 2),
+ ("streaming", 40, 51, 11, 2),
+ ("spark", 40, 50, 10, 1),
+ ("structured", 41, 51, 10, 1)
+ ),
+ StopStream
+ )
+
+ val statePath = new Path(checkpointDir.toString, "state/0")
+ val operatorMetadata = new OperatorStateMetadataReader(statePath,
hadoopConf).read()
+ .asInstanceOf[OperatorStateMetadataV1]
+
+ val expectedMetadata = OperatorStateMetadataV1(
+ OperatorInfoV1(0, "sessionWindowStateStoreSaveExec"),
+ Array(StateStoreMetadataV1("default", 1,
spark.sessionState.conf.numShufflePartitions))
+ )
+ assert(sameOperatorStateMetadata(operatorMetadata, expectedMetadata))
+ }
Review Comment:
Let's add another test to verify multiple stateful operators (opId being 0,
1, 2, etc.).
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
