[SPARK-19447] Make Range operator generate "recordsRead" metric

## What changes were proposed in this pull request?

The Range was modified to produce "recordsRead" metric instead of "generated rows". The tests were updated and partially moved to SQLMetricsSuite.

## How was this patch tested?

Unit tests.

Author: Ala Luszczak <[email protected]>

Closes #16960 from ala/range-records-read.

Author: ala

sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/codegen/CodeGenerator.scala

@@ -31,6 +31,7 @@ import org.codehaus.janino.{ByteArrayClassLoader, ClassBodyEvaluator, SimpleComp
 import org.codehaus.janino.util.ClassFile
 
 import org.apache.spark.{SparkEnv, TaskContext, TaskKilledException}
+import org.apache.spark.executor.InputMetrics
 import org.apache.spark.internal.Logging
 import org.apache.spark.metrics.source.CodegenMetrics
 import org.apache.spark.sql.catalyst.InternalRow
@@ -933,7 +934,8 @@ object CodeGenerator extends Logging {
       classOf[UnsafeMapData].getName,
       classOf[Expression].getName,
       classOf[TaskContext].getName,
-      classOf[TaskKilledException].getName
+      classOf[TaskKilledException].getName,
+      classOf[InputMetrics].getName
     ))
     evaluator.setExtendedClass(classOf[GeneratedClass])
 

sql/core/src/main/scala/org/apache/spark/sql/execution/basicPhysicalOperators.scala

@@ -365,6 +365,9 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
 
     val taskContext = ctx.freshName("taskContext")
     ctx.addMutableState("TaskContext", taskContext, s"$taskContext = TaskContext.get();")
+    val inputMetrics = ctx.freshName("inputMetrics")
+    ctx.addMutableState("InputMetrics", inputMetrics,
+        s"$inputMetrics = $taskContext.taskMetrics().inputMetrics();")
 
     // In order to periodically update the metrics without inflicting performance penalty, this
     // operator produces elements in batches. After a batch is complete, the metrics are updated
@@ -460,7 +463,7 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
       |     if ($nextBatchTodo == 0) break;
       |   }
       |   $numOutput.add($nextBatchTodo);
-      |   $numGenerated.add($nextBatchTodo);
+      |   $inputMetrics.incRecordsRead($nextBatchTodo);
       |
       |   $batchEnd += $nextBatchTodo * ${step}L;
       | }
@@ -469,7 +472,6 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
 
   protected override def doExecute(): RDD[InternalRow] = {
     val numOutputRows = longMetric("numOutputRows")
-    val numGeneratedRows = longMetric("numGeneratedRows")
     sqlContext
       .sparkContext
       .parallelize(0 until numSlices, numSlices)
@@ -488,10 +490,12 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
         val safePartitionEnd = getSafeMargin(partitionEnd)
         val rowSize = UnsafeRow.calculateBitSetWidthInBytes(1) + LongType.defaultSize
         val unsafeRow = UnsafeRow.createFromByteArray(rowSize, 1)
+        val taskContext = TaskContext.get()
 
         val iter = new Iterator[InternalRow] {
           private[this] var number: Long = safePartitionStart
           private[this] var overflow: Boolean = false
+          private[this] val inputMetrics = taskContext.taskMetrics().inputMetrics
 
           override def hasNext =
             if (!overflow) {
@@ -513,12 +517,12 @@ case class RangeExec(range: org.apache.spark.sql.catalyst.plans.logical.Range)
             }
 
             numOutputRows += 1
-            numGeneratedRows += 1
+            inputMetrics.incRecordsRead(1)
             unsafeRow.setLong(0, ret)
             unsafeRow
           }
         }
-        new InterruptibleIterator(TaskContext.get(), iter)
+        new InterruptibleIterator(taskContext, iter)
       }
   }
 

sql/core/src/test/scala/org/apache/spark/sql/execution/InputGeneratedOutputMetricsSuite.scala

@@ -17,7 +17,12 @@
 
 package org.apache.spark.sql.execution.metric
 
+import java.io.File
+
+import scala.collection.mutable.HashMap
+
 import org.apache.spark.SparkFunSuite
+import org.apache.spark.scheduler.{SparkListener, SparkListenerTaskEnd}
 import org.apache.spark.sql._
 import org.apache.spark.sql.catalyst.plans.logical.LocalRelation
 import org.apache.spark.sql.execution.SparkPlanInfo
@@ -309,4 +314,103 @@ class SQLMetricsSuite extends SparkFunSuite with SharedSQLContext {
     assert(metricInfoDeser.metadata === Some(AccumulatorContext.SQL_ACCUM_IDENTIFIER))
   }
 
+  test("range metrics") {
+    val res1 = InputOutputMetricsHelper.run(
+      spark.range(30).filter(x => x % 3 == 0).toDF()
+    )
+    assert(res1 === (30L, 0L, 30L) :: Nil)
+
+    val res2 = InputOutputMetricsHelper.run(
+      spark.range(150).repartition(4).filter(x => x < 10).toDF()
+    )
+    assert(res2 === (150L, 0L, 150L) :: (0L, 150L, 10L) :: Nil)
+
+    withTempDir { tempDir =>
+      val dir = new File(tempDir, "pqS").getCanonicalPath
+
+      spark.range(10).write.parquet(dir)
+      spark.read.parquet(dir).createOrReplaceTempView("pqS")
+
+      val res3 = InputOutputMetricsHelper.run(
+        spark.range(30).repartition(3).crossJoin(sql("select * from pqS")).repartition(2).toDF()
+      )
+      // The query above is executed in the following stages:
+      //   1. sql("select * from pqS")    => (10, 0, 10)
+      //   2. range(30)                   => (30, 0, 30)
+      //   3. crossJoin(...) of 1. and 2. => (0, 30, 300)
+      //   4. shuffle & return results    => (0, 300, 0)
+      assert(res3 === (10L, 0L, 10L) :: (30L, 0L, 30L) :: (0L, 30L, 300L) :: (0L, 300L, 0L) :: Nil)
+    }
+  }
+}
+
+object InputOutputMetricsHelper {
+  private class InputOutputMetricsListener extends SparkListener {
+    private case class MetricsResult(
+        var recordsRead: Long = 0L,
+        var shuffleRecordsRead: Long = 0L,
+        var sumMaxOutputRows: Long = 0L)
+
+    private[this] val stageIdToMetricsResult = HashMap.empty[Int, MetricsResult]
+
+    def reset(): Unit = {
+      stageIdToMetricsResult.clear()
+    }
+
+    /**
+     * Return a list of recorded metrics aggregated per stage.
+     *
+     * The list is sorted in the ascending order on the stageId.
+     * For each recorded stage, the following tuple is returned:
+     *  - sum of inputMetrics.recordsRead for all the tasks in the stage
+     *  - sum of shuffleReadMetrics.recordsRead for all the tasks in the stage
+     *  - sum of the highest values of "number of output rows" metric for all the tasks in the stage
+     */
+    def getResults(): List[(Long, Long, Long)] = {
+      stageIdToMetricsResult.keySet.toList.sorted.map { stageId =>
+        val res = stageIdToMetricsResult(stageId)
+        (res.recordsRead, res.shuffleRecordsRead, res.sumMaxOutputRows)
+      }
+    }
+
+    override def onTaskEnd(taskEnd: SparkListenerTaskEnd): Unit = synchronized {
+      val res = stageIdToMetricsResult.getOrElseUpdate(taskEnd.stageId, MetricsResult())
+
+      res.recordsRead += taskEnd.taskMetrics.inputMetrics.recordsRead
+      res.shuffleRecordsRead += taskEnd.taskMetrics.shuffleReadMetrics.recordsRead
+
+      var maxOutputRows = 0L
+      for (accum <- taskEnd.taskMetrics.externalAccums) {
+        val info = accum.toInfo(Some(accum.value), None)
+        if (info.name.toString.contains("number of output rows")) {
+          info.update match {
+            case Some(n: Number) =>
+              if (n.longValue() > maxOutputRows) {
+                maxOutputRows = n.longValue()
+              }
+            case _ => // Ignore.
+          }
+        }
+      }
+      res.sumMaxOutputRows += maxOutputRows
+    }
+  }
+
+  // Run df.collect() and return aggregated metrics for each stage.
+  def run(df: DataFrame): List[(Long, Long, Long)] = {
+    val spark = df.sparkSession
+    val sparkContext = spark.sparkContext
+    val listener = new InputOutputMetricsListener()
+    sparkContext.addSparkListener(listener)
+
+    try {
+      sparkContext.listenerBus.waitUntilEmpty(5000)
+      listener.reset()
+      df.collect()
+      sparkContext.listenerBus.waitUntilEmpty(5000)
+    } finally {
+      sparkContext.removeSparkListener(listener)
+    }
+    listener.getResults()
+  }
 }

sql/core/src/test/scala/org/apache/spark/sql/execution/metric/SQLMetricsSuite.scala

@@ -31,7 +31,7 @@ import org.apache.spark.sql.execution.DataSourceScanExec
 import org.apache.spark.sql.execution.command.ExplainCommand
 import org.apache.spark.sql.execution.datasources.LogicalRelation
 import org.apache.spark.sql.execution.datasources.jdbc.{JDBCOptions, JDBCRDD, JDBCRelation, JdbcUtils}
-import org.apache.spark.sql.execution.MetricsTestHelper
+import org.apache.spark.sql.execution.metric.InputOutputMetricsHelper
 import org.apache.spark.sql.sources._
 import org.apache.spark.sql.test.SharedSQLContext
 import org.apache.spark.sql.types._
@@ -917,13 +917,10 @@ class JDBCSuite extends SparkFunSuite
     assert(e2.contains("User specified schema not supported with `jdbc`"))
   }
 
-  test("Input/generated/output metrics on JDBC") {
+  test("Checking metrics correctness with JDBC") {
     val foobarCnt = spark.table("foobar").count()
-    val res = MetricsTestHelper.runAndGetMetrics(sql("SELECT * FROM foobar").toDF())
-    assert(res.recordsRead === foobarCnt :: Nil)
-    assert(res.shuffleRecordsRead.sum === 0)
-    assert(res.generatedRows.isEmpty)
-    assert(res.outputRows === foobarCnt :: Nil)
+    val res = InputOutputMetricsHelper.run(sql("SELECT * FROM foobar").toDF())
+    assert(res === (foobarCnt, 0L, foobarCnt) :: Nil)
   }
 
   test("SPARK-19318: Connection properties keys should be case-sensitive.") {

sql/core/src/test/scala/org/apache/spark/sql/jdbc/JDBCSuite.scala

@@ -19,7 +19,7 @@ package org.apache.spark.sql.hive.execution
 
 import org.scalatest.BeforeAndAfterAll
 
-import org.apache.spark.sql.execution.MetricsTestHelper
+import org.apache.spark.sql.execution.metric.InputOutputMetricsHelper
 import org.apache.spark.sql.hive.test.TestHive
 
 /**
@@ -49,21 +49,15 @@ class HiveSerDeSuite extends HiveComparisonTest with BeforeAndAfterAll {
 
   createQueryTest("Read Partitioned with AvroSerDe", "SELECT * FROM episodes_part")
 
-  test("Test input/generated/output metrics") {
+  test("Checking metrics correctness") {
     import TestHive._
 
     val episodesCnt = sql("select * from episodes").count()
-    val episodesRes = MetricsTestHelper.runAndGetMetrics(sql("select * from episodes").toDF())
-    assert(episodesRes.recordsRead === episodesCnt :: Nil)
-    assert(episodesRes.shuffleRecordsRead.sum === 0)
-    assert(episodesRes.generatedRows.isEmpty)
-    assert(episodesRes.outputRows === episodesCnt :: Nil)
+    val episodesRes = InputOutputMetricsHelper.run(sql("select * from episodes").toDF())
+    assert(episodesRes === (episodesCnt, 0L, episodesCnt) :: Nil)
 
     val serdeinsCnt = sql("select * from serdeins").count()
-    val serdeinsRes = MetricsTestHelper.runAndGetMetrics(sql("select * from serdeins").toDF())
-    assert(serdeinsRes.recordsRead === serdeinsCnt :: Nil)
-    assert(serdeinsRes.shuffleRecordsRead.sum === 0)
-    assert(serdeinsRes.generatedRows.isEmpty)
-    assert(serdeinsRes.outputRows === serdeinsCnt :: Nil)
+    val serdeinsRes = InputOutputMetricsHelper.run(sql("select * from serdeins").toDF())
+    assert(serdeinsRes === (serdeinsCnt, 0L, serdeinsCnt) :: Nil)
   }
 }