[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r467319212
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r467318625
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r467318270
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325: URL: https://github.com/apache/hive/pull/1325#discussion_r466421484 ## File path: ql/src/test/results/clientpositive/llap/dynamic_semijoin_reduction_2.q.out ## @@ -79,27 +79,25 @@ STAGE PLANS: Tez A masked pattern was here Edges: Review comment: The plan in `dynamic_semijoin_reduction_2.q` has three single column semijoin reducers that get merged to one multi column one. As a result three reducers get merged to one thus making the plan more compact. Apart from that, you are right that the multi column transformation can lead to further optimization opportunities. An example can be seen in query24.q.out (Check commit https://github.com/apache/hive/pull/1325/commits/c9f9112d0802906dce7442f3d4c01535a584af11). There the `SharedWorkOptimizer` kicks in and merges two semijoin reducer branches on the same scan operator. 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org For additional commands, e-mail: gitbox-h...@hive.apache.org
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466414395
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466410669
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325: URL: https://github.com/apache/hive/pull/1325#discussion_r466404775 ## File path: ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java ## @@ -2054,7 +2067,8 @@ private void markSemiJoinForDPP(OptimizeTezProcContext procCtx) // Lookup nDVs on TS side. RuntimeValuesInfo rti = procCtx.parseContext .getRsToRuntimeValuesInfoMap().get(rs); - ExprNodeDesc tsExpr = rti.getTsColExpr(); + // TODO Adapt for multi column semi-joins. Review comment: I meant to handle this as part of HIVE-23934. I added the reference to the comment. 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org For additional commands, e-mail: gitbox-h...@hive.apache.org
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325: URL: https://github.com/apache/hive/pull/1325#discussion_r466404256 ## File path: ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java ## @@ -1887,13 +1898,14 @@ private void removeSemijoinOptimizationByBenefit(OptimizeTezProcContext procCtx) // Check the ndv/rows from the SEL vs the destination tablescan the semijoin opt is going to. TableScanOperator ts = sjInfo.getTsOp(); RuntimeValuesInfo rti = procCtx.parseContext.getRsToRuntimeValuesInfoMap().get(rs); -ExprNodeDesc tsExpr = rti.getTsColExpr(); -// In the SEL operator of the semijoin branch, there should be only one column in the operator -ExprNodeDesc selExpr = sel.getConf().getColList().get(0); +List targetColumns = rti.getTargetColumns(); +// In multi column semijoin branches the last column of the SEL operator is hash(c1, c2, ..., cn) +// so we shouldn't consider it. +List sourceColumns = sel.getConf().getColList().subList(0, targetColumns.size()); Review comment: Done 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org - To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org For additional commands, e-mail: gitbox-h...@hive.apache.org
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466368067
##
File path: ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java
##
@@ -1737,35 +1737,46 @@ private static double getBloomFilterBenefit(
}
}
-// Selectivity: key cardinality of semijoin / domain cardinality
-// Benefit (rows filtered from ts): (1 - selectivity) * # ts rows
-double selectivity = selKeyCardinality / (double) keyDomainCardinality;
-selectivity = Math.min(selectivity, 1);
-benefit = tsRows * (1 - selectivity);
-
if (LOG.isDebugEnabled()) {
- LOG.debug("BloomFilter benefit for " + selCol + " to " + tsCol
- + ", selKeyCardinality=" + selKeyCardinality
- + ", tsKeyCardinality=" + tsKeyCardinality
- + ", tsRows=" + tsRows
- + ", keyDomainCardinality=" + keyDomainCardinality);
- LOG.debug("SemiJoin key selectivity=" + selectivity
- + ", benefit=" + benefit);
+ LOG.debug("BloomFilter selectivity for " + selCol + " to " + tsCol + ",
selKeyCardinality=" + selKeyCardinality
+ + ", tsKeyCardinality=" + tsKeyCardinality + ",
keyDomainCardinality=" + keyDomainCardinality);
}
+// Selectivity: key cardinality of semijoin / domain cardinality
+return selKeyCardinality / (double) keyDomainCardinality;
+ }
-return benefit;
+ private static double getBloomFilterBenefit(
+ SelectOperator sel, List selExpr,
+ Statistics filStats, List tsExpr) {
+if (sel.getStatistics() == null || filStats == null) {
+ LOG.debug("No stats available to compute BloomFilter benefit");
+ return -1;
+}
+double selectivity = 0.0;
+for (int i = 0; i < tsExpr.size(); i++) {
+ selectivity = Math.max(selectivity, getBloomFilterSelectivity(sel,
selExpr.get(i), filStats, tsExpr.get(i)));
Review comment:
You are right, I was the one confused. I applied the change along with
some doc.
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
-
To unsubscribe, e-mail: gitbox-unsubscr...@hive.apache.org
For additional commands, e-mail: gitbox-h...@hive.apache.org
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466367481
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466366197
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466365583
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466365453
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
Review comment:
Done. Didn't add `checkNotNull` since NPE will be thrown anyways and it
is rather informative as well.
This is an automated message from the Apache Git Service.
To respond
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r466347783
##
File path: ql/src/java/org/apache/hadoop/hive/ql/exec/OperatorUtils.java
##
@@ -53,6 +53,34 @@
private static final Logger LOG =
LoggerFactory.getLogger(OperatorUtils.class);
+ /**
+ * Return the ancestor of the specified operator at the provided path or
null if the path is invalid.
+ *
+ * The method is equivalent to following code:
+ * {@code
+ * op.getParentOperators().get(path[0])
+ * .getParentOperators().get(path[1])
+ * ...
+ * .getParentOperators().get(path[n])
+ * }
+ * with additional checks about the validity of the provided path and the
type of the ancestor.
+ *
+ * @param op the operator for which we
+ * @param clazz the class of the ancestor operator
+ * @param path the path leading to the desired ancestor
+ * @param the type of the ancestor
+ * @return the ancestor of the specified operator at the provided path or
null if the path is invalid.
+ */
+ public static T ancestor(Operator op, Class clazz, int... path) {
+Operator target = op;
+for (int i = 0; i < path.length; i++) {
+ if (target.getParentOperators() == null || path[i] >
target.getParentOperators().size())
Review comment:
Done. I also configured IntelliJ to force their usage in single line
statements so hopefully they should never appear.
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r465207659
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r465206179
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
+ assert selOp.getParentOperators().size() == 1;
+ Operator source = selOp.getParentOperators().get(0);
+ SJSourceTarget sjKey = new SJSourceTarget(source, ts);
+ List ops = sameTableSJ.computeIfAbsent(sjKey,
tableScanOperator -> new
[GitHub] [hive] zabetak commented on a change in pull request #1325: HIVE-21196 HIVE-23940 Multi-column semijoin reducers and TPC-H datasets
zabetak commented on a change in pull request #1325:
URL: https://github.com/apache/hive/pull/1325#discussion_r465192529
##
File path:
ql/src/java/org/apache/hadoop/hive/ql/optimizer/SemiJoinReductionMerge.java
##
@@ -0,0 +1,399 @@
+/*
+ * 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.hadoop.hive.ql.optimizer;
+
+import org.apache.hadoop.hive.conf.HiveConf;
+import org.apache.hadoop.hive.ql.exec.ColumnInfo;
+import org.apache.hadoop.hive.ql.exec.FilterOperator;
+import org.apache.hadoop.hive.ql.exec.GroupByOperator;
+import org.apache.hadoop.hive.ql.exec.Operator;
+import org.apache.hadoop.hive.ql.exec.OperatorFactory;
+import org.apache.hadoop.hive.ql.exec.OperatorUtils;
+import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
+import org.apache.hadoop.hive.ql.exec.RowSchema;
+import org.apache.hadoop.hive.ql.exec.SelectOperator;
+import org.apache.hadoop.hive.ql.exec.TableScanOperator;
+import org.apache.hadoop.hive.ql.exec.Utilities;
+import org.apache.hadoop.hive.ql.io.AcidUtils;
+import org.apache.hadoop.hive.ql.parse.GenTezUtils;
+import org.apache.hadoop.hive.ql.parse.ParseContext;
+import org.apache.hadoop.hive.ql.parse.RuntimeValuesInfo;
+import org.apache.hadoop.hive.ql.parse.SemanticAnalyzer;
+import org.apache.hadoop.hive.ql.parse.SemanticException;
+import org.apache.hadoop.hive.ql.parse.SemiJoinBranchInfo;
+import org.apache.hadoop.hive.ql.plan.AggregationDesc;
+import org.apache.hadoop.hive.ql.plan.DynamicValue;
+import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeDynamicValueDesc;
+import org.apache.hadoop.hive.ql.plan.ExprNodeGenericFuncDesc;
+import org.apache.hadoop.hive.ql.plan.FilterDesc;
+import org.apache.hadoop.hive.ql.plan.GroupByDesc;
+import org.apache.hadoop.hive.ql.plan.PlanUtils;
+import org.apache.hadoop.hive.ql.plan.ReduceSinkDesc;
+import org.apache.hadoop.hive.ql.plan.SelectDesc;
+import org.apache.hadoop.hive.ql.plan.TableDesc;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFEvaluator;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMax;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDAFMin;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFBetween;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFInBloomFilter;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFMurmurHash;
+import org.apache.hadoop.hive.ql.udf.generic.GenericUDFOPAnd;
+import org.apache.hadoop.hive.ql.util.NullOrdering;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfo;
+import org.apache.hadoop.hive.serde2.typeinfo.TypeInfoFactory;
+
+import java.util.ArrayDeque;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.Comparator;
+import java.util.Deque;
+import java.util.EnumSet;
+import java.util.HashMap;
+import java.util.List;
+import java.util.Map;
+import java.util.SortedMap;
+import java.util.TreeMap;
+
+public class SemiJoinReductionMerge extends Transform {
+
+ public ParseContext transform(ParseContext parseContext) throws
SemanticException {
+Map map =
parseContext.getRsToSemiJoinBranchInfo();
+if (map.isEmpty()) {
+ return parseContext;
+}
+HiveConf hiveConf = parseContext.getConf();
+
+// Order does not really matter but it is necessary to keep plans stable
+SortedMap> sameTableSJ =
+new TreeMap<>(Comparator.comparing(SJSourceTarget::toString));
+for (Map.Entry smjEntry :
map.entrySet()) {
+ TableScanOperator ts = smjEntry.getValue().getTsOp();
+ // Semijoin optimization branch should look like
-SEL-GB1-RS1-GB2-RS2
+ SelectOperator selOp = OperatorUtils.ancestor(smjEntry.getKey(),
SelectOperator.class, 0, 0, 0, 0);
+ assert selOp != null;
Review comment:
I was having the same internal debate with my self and in the end I
concluded to the following. If we arrive here then something went terribly
wrong so it's better to fail miserably rather than throwing an exception that
someone might
