choose join order by connectivity

englefly · Nov 30, 2023 · 377a7be · 377a7be
1 parent ce70353
commit 377a7be
Show file tree

Hide file tree

Showing 5 changed files with 39 additions and 1 deletion.
diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/PlanContext.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/PlanContext.java
@@ -88,4 +88,8 @@ public Statistics getChildStatistics(int index) {
     public List<Statistics> getChildrenStatistics() {
         return childrenStats;
     }
+
+    public StatementContext getStatementContext() {
+        return connectContext.getStatementContext();
+    }
 }
diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/StatementContext.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/StatementContext.java
@@ -42,6 +42,8 @@
 import com.google.common.collect.Maps;
 import com.google.common.collect.Sets;
 
+import java.util.ArrayList;
+import java.util.Collection;
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
@@ -90,6 +92,8 @@ public class StatementContext {
     private final Map<String, Hint> hintMap = Maps.newLinkedHashMap();
     private final Set<String> viewDdlSqlSet = Sets.newHashSet();
 
+    private final List<Expression> joinFilters = new ArrayList<>();
+
     public StatementContext() {
         this.connectContext = ConnectContext.get();
     }
@@ -246,4 +250,12 @@ public void addViewDdlSql(String ddlSql) {
     public List<String> getViewDdlSqls() {
         return ImmutableList.copyOf(viewDdlSqlSet);
     }
+
+    public List<Expression> getJoinFilters() {
+        return joinFilters;
+    }
+
+    public void addJoinFilters(Collection<Expression> newJoinFilters) {
+        this.joinFilters.addAll(newJoinFilters);
+    }
 }
diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/cost/CostModelV1.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/cost/CostModelV1.java
@@ -22,6 +22,7 @@
 import org.apache.doris.nereids.properties.DistributionSpecGather;
 import org.apache.doris.nereids.properties.DistributionSpecHash;
 import org.apache.doris.nereids.properties.DistributionSpecReplicated;
+import org.apache.doris.nereids.trees.expressions.Expression;
 import org.apache.doris.nereids.trees.plans.Plan;
 import org.apache.doris.nereids.trees.plans.physical.PhysicalAssertNumRows;
 import org.apache.doris.nereids.trees.plans.physical.PhysicalDeferMaterializeOlapScan;
@@ -48,6 +49,8 @@
 
 import com.google.common.base.Preconditions;
 
+import java.util.Collections;
+
 class CostModelV1 extends PlanVisitor<Cost, PlanContext> {
 
     // for a join, skew = leftRowCount/rightRowCount
@@ -262,6 +265,15 @@ public Cost visitPhysicalHashJoin(
 
         double leftRowCount = probeStats.getRowCount();
         double rightRowCount = buildStats.getRowCount();
+        if (Math.abs(leftRowCount - rightRowCount) < 10) {
+            int leftConnectivity = computeConnectivity(physicalHashJoin.left(), context);
+            int rightConnectivity = computeConnectivity(physicalHashJoin.right(), context);
+            if (rightConnectivity < leftConnectivity) {
+                leftRowCount += 10;
+            }
+
+        }
+
         /*
         pattern1: L join1 (Agg1() join2 Agg2())
         result number of join2 may much less than Agg1.
@@ -310,6 +322,15 @@ public Cost visitPhysicalHashJoin(
         );
     }
 
+    private int computeConnectivity(
+            Plan plan, PlanContext context) {
+        int connectCount = 0;
+        for (Expression expr : context.getStatementContext().getJoinFilters()) {
+            connectCount += Collections.disjoint(expr.getInputSlots(), plan.getOutputSet()) ? 0 : 1;
+        }
+        return connectCount;
+    }
+
     @Override
     public Cost visitPhysicalNestedLoopJoin(
             PhysicalNestedLoopJoin<? extends Plan, ? extends Plan> nestedLoopJoin,

diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/exploration/join/JoinCommute.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/exploration/join/JoinCommute.java
@@ -93,7 +93,7 @@ public static boolean check(SwapType swapType, LogicalJoin<GroupPlan, GroupPlan>
         if (swapType == SwapType.LEFT_ZIG_ZAG) {
             double leftRows = join.left().getGroup().getStatistics().getRowCount();
             double rightRows = join.right().getGroup().getStatistics().getRowCount();
-            return leftRows < rightRows && isZigZagJoin(join);
+            return leftRows <= rightRows && isZigZagJoin(join);
         }
 
         return true;

diff --git a/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/rewrite/ReorderJoin.java b/fe/fe-core/src/main/java/org/apache/doris/nereids/rules/rewrite/ReorderJoin.java
@@ -87,6 +87,7 @@ public Rule build() {
                 Plan plan = joinToMultiJoin(filter, planToHintType);
                 Preconditions.checkState(plan instanceof MultiJoin);
                 MultiJoin multiJoin = (MultiJoin) plan;
+                ctx.statementContext.addJoinFilters(multiJoin.getJoinFilter());
                 ctx.statementContext.setMaxNAryInnerJoin(multiJoin.children().size());
                 Plan after = multiJoinToJoin(multiJoin, planToHintType);
                 return after;