From 8dfd6cfa78f55d70419075a3f3c9ad64b8daed05 Mon Sep 17 00:00:00 2001
From: PHILO-HE <feilong.he@intel.com>
Date: Mon, 30 Oct 2023 21:22:30 +0800
Subject: [PATCH] Refine the code

---
 .../extension/ExpandFallbackPolicy.scala      | 92 ++++++++++++++-----
 1 file changed, 70 insertions(+), 22 deletions(-)

diff --git a/gluten-core/src/main/scala/io/glutenproject/extension/ExpandFallbackPolicy.scala b/gluten-core/src/main/scala/io/glutenproject/extension/ExpandFallbackPolicy.scala
index 191c4a80f1d15..2c5f0825647ac 100644
--- a/gluten-core/src/main/scala/io/glutenproject/extension/ExpandFallbackPolicy.scala
+++ b/gluten-core/src/main/scala/io/glutenproject/extension/ExpandFallbackPolicy.scala
@@ -67,49 +67,93 @@ import org.apache.spark.sql.execution.exchange.{Exchange, ReusedExchangeExec}
 case class ExpandFallbackPolicy(isAdaptiveContext: Boolean, originalPlan: SparkPlan)
   extends Rule[SparkPlan] {
 
-  private def countColumnarToRowWhenFallbackStage(plan: SparkPlan): Int = {
-    plan
-      .collectLeaves()
-      .filter(
-        p =>
-          p match {
-            case q: QueryStageExec if q.supportsColumnar => true
-            case _ => false
-          })
-      .size
-  }
-
-  private def countFallbacks(plan: SparkPlan): Int = {
+  private def countFallback(plan: SparkPlan): Int = {
     var fallbacks = 0
-    def countFallback(plan: SparkPlan): Unit = {
+    def countFallbackInternal(plan: SparkPlan): Unit = {
       plan match {
         case _: QueryStageExec => // Another stage.
         case _: CommandResultExec | _: ExecutedCommandExec => // ignore
         // we plan exchange to columnar exchange in columnar rules and the exchange does not
         // support columnar, so the output columnar is always false in AQE postStageCreationRules
         case ColumnarToRowExec(s: Exchange) if isAdaptiveContext =>
-          countFallback(s)
+          countFallbackInternal(s)
         case u: UnaryExecNode
             if !u.isInstanceOf[GlutenPlan] && InMemoryTableScanHelper.isGlutenTableCache(u.child) =>
           // Vanilla Spark plan will call `InMemoryTableScanExec.convertCachedBatchToInternalRow`
           // which is a kind of `ColumnarToRowExec`.
           fallbacks = fallbacks + 1
-          countFallback(u.child)
+          countFallbackInternal(u.child)
         case ColumnarToRowExec(p: GlutenPlan) =>
           logDebug(s"Find a columnar to row for gluten plan:\n$p")
           fallbacks = fallbacks + 1
-          countFallback(p)
+          countFallbackInternal(p)
         case leafPlan: LeafExecNode if InMemoryTableScanHelper.isGlutenTableCache(leafPlan) =>
         case leafPlan: LeafExecNode if !leafPlan.isInstanceOf[GlutenPlan] =>
           // Possible fallback for leaf node.
           fallbacks = fallbacks + 1
-        case p => p.children.foreach(countFallback)
+        case p => p.children.foreach(countFallbackInternal)
       }
     }
-    countFallback(plan)
+    countFallbackInternal(plan)
     fallbacks
   }
 
+  /**
+   * When making a stage fall back, it's possible that we need a ColumnarToRow to adapt to last
+   * stage's columnar output. So we need to evaluate the cost, i.e., the number of required
+   * ColumnarToRow between entirely fallback stage and last stage(s). Thus, we can avoid possible
+   * performance degradation caused by fallback policy.
+   *
+   * spotless:off
+   *
+   * Spark plan before applying fallback policy:
+   *
+   *        ColumnarExchange
+   *  ----------- | --------------- last stage
+   *    HashAggregateTransformer
+   *              |
+   *        ColumnarToRow
+   *              |
+   *           Project
+   *
+   * To illustrate the effect if cost is not taken into account, here is spark plan
+   * after applying whole stage fallback policy (threshold = 1):
+   *
+   *        ColumnarExchange
+   *  ----------- | --------------- last stage
+   *         ColumnarToRow
+   *              |
+   *         HashAggregate
+   *              |
+   *           Project
+   *
+   *  So by considering the cost, we will not apply the fallback policy.
+   *
+   * spotless:on
+   */
+  private def countStageFallbackCost(plan: SparkPlan): Int = {
+    var stageFallbackCost = 0
+
+    /**
+     * Find a Gluten plan whose child is QueryStageExec. Then, increase stageFallbackCost if the
+     * last query stage's output is columnar.
+     */
+    def countStageFallbackCostInternal(plan: SparkPlan): Unit = {
+      plan match {
+        case p: GlutenPlan if p.children.find(_.isInstanceOf[QueryStageExec]).isDefined =>
+          p.children
+            .filter(_.isInstanceOf[QueryStageExec])
+            .foreach {
+              case stage: QueryStageExec if stage.supportsColumnar =>
+                stageFallbackCost = stageFallbackCost + 1
+            }
+        case p => p.children.foreach(countStageFallbackCostInternal)
+      }
+    }
+    countStageFallbackCostInternal(plan)
+    stageFallbackCost
+  }
+
   private def hasColumnarBroadcastExchangeWithJoin(plan: SparkPlan): Boolean = {
     def isColumnarBroadcastExchange(p: SparkPlan): Boolean = p match {
       case BroadcastQueryStageExec(_, _: ColumnarBroadcastExchangeExec, _) => true
@@ -146,11 +190,15 @@ case class ExpandFallbackPolicy(isAdaptiveContext: Boolean, originalPlan: SparkP
       return None
     }
 
-    val netFallbackNum = countFallbacks(plan) - countColumnarToRowWhenFallbackStage(plan)
+    val netFallbackNum = if (isAdaptiveContext) {
+      countFallback(plan) - countStageFallbackCost(plan)
+    } else {
+      countFallback(plan)
+    }
     if (netFallbackNum >= fallbackThreshold) {
       Some(
-        s"Fall back the plan due to net fallback number $netFallbackNum, " +
-          s"threshold $fallbackThreshold")
+        s"Fallback policy is taking effect, net fallback number: $netFallbackNum, " +
+          s"threshold: $fallbackThreshold")
     } else {
       None
     }