[SPARK-1113] External spilling - fix Int.MaxValue hash code collision bug

The original poster of this bug is @guojc, who opened a PR that preceded this one at https://github.com/apache/incubator-spark/pull/612.

ExternalAppendOnlyMap uses key hash code to order the buffer streams from which spilled files are read back into memory. When a buffer stream is empty, the default hash code for that stream is equal to Int.MaxValue. This is, however, a perfectly legitimate candidate for a key hash code. When reading from a spilled map containing such a key, a hash collision may occur, in which case we attempt to read from an empty stream and throw NoSuchElementException.

The fix is to maintain the invariant that empty buffer streams are never added back to the merge queue to be considered. This guarantees that we never read from an empty buffer stream, ever again.

This PR also includes two new tests for hash collisions.

Author: Andrew Or <andrewor14@gmail.com>

Closes #624 from andrewor14/spilling-bug and squashes the following commits:

9e7263d [Andrew Or] Slightly optimize next()
2037ae2 [Andrew Or] Move a few comments around...
cf95942 [Andrew Or] Remove default value of Int.MaxValue for minKeyHash
c11f03b [Andrew Or] Fix Int.MaxValue hash collision bug in ExternalAppendOnlyMap
21c1a39 [Andrew Or] Add hash collision tests to ExternalAppendOnlyMapSuite

core/src/main/scala/org/apache/spark/util/collection/ExternalAppendOnlyMap.scala

@@ -148,7 +148,7 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
   }
 
   /**
-   * Sort the existing contents of the in-memory map and spill them to a temporary file on disk
+   * Sort the existing contents of the in-memory map and spill them to a temporary file on disk.
    */
   private def spill(mapSize: Long) {
     spillCount += 1
@@ -223,7 +223,8 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
    */
   private class ExternalIterator extends Iterator[(K, C)] {
 
-    // A fixed-size queue that maintains a buffer for each stream we are currently merging
+    // A queue that maintains a buffer for each stream we are currently merging
+    // This queue maintains the invariant that it only contains non-empty buffers
     private val mergeHeap = new mutable.PriorityQueue[StreamBuffer]
 
     // Input streams are derived both from the in-memory map and spilled maps on disk
@@ -233,7 +234,9 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
 
     inputStreams.foreach { it =>
       val kcPairs = getMorePairs(it)
-      mergeHeap.enqueue(StreamBuffer(it, kcPairs))
+      if (kcPairs.length > 0) {
+        mergeHeap.enqueue(new StreamBuffer(it, kcPairs))
+      }
     }
 
     /**
@@ -258,11 +261,11 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
 
     /**
      * If the given buffer contains a value for the given key, merge that value into
-     * baseCombiner and remove the corresponding (K, C) pair from the buffer
+     * baseCombiner and remove the corresponding (K, C) pair from the buffer.
      */
     private def mergeIfKeyExists(key: K, baseCombiner: C, buffer: StreamBuffer): C = {
       var i = 0
-      while (i < buffer.pairs.size) {
+      while (i < buffer.pairs.length) {
         val (k, c) = buffer.pairs(i)
         if (k == key) {
           buffer.pairs.remove(i)
@@ -274,40 +277,41 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
     }
 
     /**
-     * Return true if there exists an input stream that still has unvisited pairs
+     * Return true if there exists an input stream that still has unvisited pairs.
      */
-    override def hasNext: Boolean = mergeHeap.exists(!_.pairs.isEmpty)
+    override def hasNext: Boolean = mergeHeap.length > 0
 
     /**
      * Select a key with the minimum hash, then combine all values with the same key from all
      * input streams.
      */
     override def next(): (K, C) = {
+      if (mergeHeap.length == 0) {
+        throw new NoSuchElementException
+      }
       // Select a key from the StreamBuffer that holds the lowest key hash
       val minBuffer = mergeHeap.dequeue()
       val (minPairs, minHash) = (minBuffer.pairs, minBuffer.minKeyHash)
-      if (minPairs.length == 0) {
-        // Should only happen when no other stream buffers have any pairs left
-        throw new NoSuchElementException
-      }
       var (minKey, minCombiner) = minPairs.remove(0)
       assert(minKey.hashCode() == minHash)
 
       // For all other streams that may have this key (i.e. have the same minimum key hash),
       // merge in the corresponding value (if any) from that stream
       val mergedBuffers = ArrayBuffer[StreamBuffer](minBuffer)
-      while (!mergeHeap.isEmpty && mergeHeap.head.minKeyHash == minHash) {
+      while (mergeHeap.length > 0 && mergeHeap.head.minKeyHash == minHash) {
         val newBuffer = mergeHeap.dequeue()
         minCombiner = mergeIfKeyExists(minKey, minCombiner, newBuffer)
         mergedBuffers += newBuffer
       }
 
-      // Repopulate each visited stream buffer and add it back to the merge heap
+      // Repopulate each visited stream buffer and add it back to the queue if it is non-empty
       mergedBuffers.foreach { buffer =>
-        if (buffer.pairs.length == 0) {
+        if (buffer.isEmpty) {
           buffer.pairs ++= getMorePairs(buffer.iterator)
         }
-        mergeHeap.enqueue(buffer)
+        if (!buffer.isEmpty) {
+          mergeHeap.enqueue(buffer)
+        }
       }
 
       (minKey, minCombiner)
@@ -323,13 +327,12 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
     private case class StreamBuffer(iterator: Iterator[(K, C)], pairs: ArrayBuffer[(K, C)])
       extends Comparable[StreamBuffer] {
 
-      def minKeyHash: Int = {
-        if (pairs.length > 0){
-          // pairs are already sorted by key hash
-          pairs(0)._1.hashCode()
-        } else {
-          Int.MaxValue
-        }
+      def isEmpty = pairs.length == 0
+
+      // Invalid if there are no more pairs in this stream
+      def minKeyHash = {
+        assert(pairs.length > 0)
+        pairs.head._1.hashCode()
       }
 
       override def compareTo(other: StreamBuffer): Int = {
@@ -356,7 +359,7 @@ private[spark] class ExternalAppendOnlyMap[K, V, C](
     private var objectsRead = 0
 
     /**
-     * Construct a stream that reads only from the next batch
+     * Construct a stream that reads only from the next batch.
      */
     private def nextBatchStream(): InputStream = {
       if (batchSizes.length > 0) {

core/src/test/scala/org/apache/spark/util/collection/ExternalAppendOnlyMapSuite.scala

@@ -19,21 +19,16 @@ package org.apache.spark.util.collection
 
 import scala.collection.mutable.ArrayBuffer
 
-import org.scalatest.{BeforeAndAfter, FunSuite}
+import org.scalatest.FunSuite
 
 import org.apache.spark._
 import org.apache.spark.SparkContext._
 
-class ExternalAppendOnlyMapSuite extends FunSuite with BeforeAndAfter with LocalSparkContext {
+class ExternalAppendOnlyMapSuite extends FunSuite with LocalSparkContext {
 
-  private val createCombiner: (Int => ArrayBuffer[Int]) = i => ArrayBuffer[Int](i)
-  private val mergeValue: (ArrayBuffer[Int], Int) => ArrayBuffer[Int] = (buffer, i) => {
-    buffer += i
-  }
-  private val mergeCombiners: (ArrayBuffer[Int], ArrayBuffer[Int]) => ArrayBuffer[Int] =
-    (buf1, buf2) => {
-      buf1 ++= buf2
-    }
+  private def createCombiner(i: Int) = ArrayBuffer[Int](i)
+  private def mergeValue(buffer: ArrayBuffer[Int], i: Int) = buffer += i
+  private def mergeCombiners(buf1: ArrayBuffer[Int], buf2: ArrayBuffer[Int]) = buf1 ++= buf2
 
   test("simple insert") {
     val conf = new SparkConf(false)
@@ -203,13 +198,13 @@ class ExternalAppendOnlyMapSuite extends FunSuite with BeforeAndAfter with Local
   }
 
   test("spilling") {
-    // TODO: Use SparkConf (which currently throws connection reset exception)
-    System.setProperty("spark.shuffle.memoryFraction", "0.001")
-    sc = new SparkContext("local-cluster[1,1,512]", "test")
+    val conf = new SparkConf(true)  // Load defaults, otherwise SPARK_HOME is not found
+    conf.set("spark.shuffle.memoryFraction", "0.001")
+    sc = new SparkContext("local-cluster[1,1,512]", "test", conf)
 
     // reduceByKey - should spill ~8 times
     val rddA = sc.parallelize(0 until 100000).map(i => (i/2, i))
-    val resultA = rddA.reduceByKey(math.max(_, _)).collect()
+    val resultA = rddA.reduceByKey(math.max).collect()
     assert(resultA.length == 50000)
     resultA.foreach { case(k, v) =>
       k match {
@@ -252,7 +247,73 @@ class ExternalAppendOnlyMapSuite extends FunSuite with BeforeAndAfter with Local
         case _ =>
       }
     }
+  }
+
+  test("spilling with hash collisions") {
+    val conf = new SparkConf(true)
+    conf.set("spark.shuffle.memoryFraction", "0.001")
+    sc = new SparkContext("local-cluster[1,1,512]", "test", conf)
+
+    def createCombiner(i: String) = ArrayBuffer[String](i)
+    def mergeValue(buffer: ArrayBuffer[String], i: String) = buffer += i
+    def mergeCombiners(buffer1: ArrayBuffer[String], buffer2: ArrayBuffer[String]) =
+      buffer1 ++= buffer2
+
+    val map = new ExternalAppendOnlyMap[String, String, ArrayBuffer[String]](
+      createCombiner, mergeValue, mergeCombiners)
+
+    val collisionPairs = Seq(
+      ("Aa", "BB"),                   // 2112
+      ("to", "v1"),                   // 3707
+      ("variants", "gelato"),         // -1249574770
+      ("Teheran", "Siblings"),        // 231609873
+      ("misused", "horsemints"),      // 1069518484
+      ("isohel", "epistolaries"),     // -1179291542
+      ("righto", "buzzards"),         // -931102253
+      ("hierarch", "crinolines"),     // -1732884796
+      ("inwork", "hypercatalexes"),   // -1183663690
+      ("wainages", "presentencing"),  // 240183619
+      ("trichothecenes", "locular"),  // 339006536
+      ("pomatoes", "eructation")      // 568647356
+    )
+
+    (1 to 100000).map(_.toString).foreach { i => map.insert(i, i) }
+    collisionPairs.foreach { case (w1, w2) =>
+      map.insert(w1, w2)
+      map.insert(w2, w1)
+    }
+
+    // A map of collision pairs in both directions
+    val collisionPairsMap = (collisionPairs ++ collisionPairs.map(_.swap)).toMap
+
+    // Avoid map.size or map.iterator.length because this destructively sorts the underlying map
+    var count = 0
+
+    val it = map.iterator
+    while (it.hasNext) {
+      val kv = it.next()
+      val expectedValue = ArrayBuffer[String](collisionPairsMap.getOrElse(kv._1, kv._1))
+      assert(kv._2.equals(expectedValue))
+      count += 1
+    }
+    assert(count == 100000 + collisionPairs.size * 2)
+  }
+
+  test("spilling with hash collisions using the Int.MaxValue key") {
+    val conf = new SparkConf(true)
+    conf.set("spark.shuffle.memoryFraction", "0.001")
+    sc = new SparkContext("local-cluster[1,1,512]", "test", conf)
 
-    System.clearProperty("spark.shuffle.memoryFraction")
+    val map = new ExternalAppendOnlyMap[Int, Int, ArrayBuffer[Int]](createCombiner,
+      mergeValue, mergeCombiners)
+
+    (1 to 100000).foreach { i => map.insert(i, i) }
+    map.insert(Int.MaxValue, Int.MaxValue)
+
+    val it = map.iterator
+    while (it.hasNext) {
+      // Should not throw NoSuchElementException
+      it.next()
+    }
   }
 }