/**
* 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 java.util.ArrayList;
import java.util.LinkedHashMap;
import java.util.List;
import java.util.Map;
import java.util.Stack;
import org.apache.hadoop.hive.conf.HiveConf;
import org.apache.hadoop.hive.ql.exec.GroupByOperator;
import org.apache.hadoop.hive.ql.exec.LimitOperator;
import org.apache.hadoop.hive.ql.exec.Operator;
import org.apache.hadoop.hive.ql.exec.ReduceSinkOperator;
import org.apache.hadoop.hive.ql.lib.DefaultGraphWalker;
import org.apache.hadoop.hive.ql.lib.DefaultRuleDispatcher;
import org.apache.hadoop.hive.ql.lib.Dispatcher;
import org.apache.hadoop.hive.ql.lib.GraphWalker;
import org.apache.hadoop.hive.ql.lib.Node;
import org.apache.hadoop.hive.ql.lib.NodeProcessor;
import org.apache.hadoop.hive.ql.lib.NodeProcessorCtx;
import org.apache.hadoop.hive.ql.lib.Rule;
import org.apache.hadoop.hive.ql.lib.RuleRegExp;
import org.apache.hadoop.hive.ql.parse.ParseContext;
import org.apache.hadoop.hive.ql.parse.SemanticException;
/**
* Make RS calculate top-K selection for limit clause.
* It's only works with RS for limit operation which means between RS and LITMIT,
* there should not be other operators which may change number of rows like FilterOperator.
* see {@link Operator#acceptLimitPushdown}
*
* If RS is only for limiting rows, RSHash counts row with same key separately.
* But if RS is for GBY, RSHash should forward all the rows with the same key.
*
* Legend : A(a) --> key A, value a, row A(a)
*
* If each RS in mapper tasks is forwarded rows like this
*
* MAP1(RS) : 40(a)-10(b)-30(c)-10(d)-70(e)-80(f)
* MAP2(RS) : 90(g)-80(h)-60(i)-40(j)-30(k)-20(l)
* MAP3(RS) : 40(m)-50(n)-30(o)-30(p)-60(q)-70(r)
*
* OBY or GBY makes result like this,
*
* REDUCER : 10(b,d)-20(l)-30(c,k,o,p)-40(a,j,m)-50(n)-60(i,q)-70(e,r)-80(f,h)-90(g)
* LIMIT 3 for GBY: 10(b,d)-20(l)-30(c,k,o,p)
* LIMIT 3 for OBY: 10(b,d)-20(l)
*
* with the optimization, the amount of shuffling can be reduced, making identical result
*
* For GBY,
*
* MAP1 : 40(a)-10(b)-30(c)-10(d)
* MAP2 : 40(j)-30(k)-20(l)
* MAP3 : 40(m)-50(n)-30(o)-30(p)
*
* REDUCER : 10(b,d)-20(l)-30(c,k,o,p)-40(a,j,m)-50(n)
* LIMIT 3 : 10(b,d)-20(l)-30(c,k,o,p)
*
* For OBY,
*
* MAP1 : 10(b)-30(c)-10(d)
* MAP2 : 40(j)-30(k)-20(l)
* MAP3 : 40(m)-50(n)-30(o)
*
* REDUCER : 10(b,d)-20(l)-30(c,k,o)-40(j,m)-50(n)
* LIMIT 3 : 10(b,d)-20(l)
*/
public class LimitPushdownOptimizer implements Transform {
public ParseContext transform(ParseContext pctx) throws SemanticException {
Map<Rule, NodeProcessor> opRules = new LinkedHashMap<Rule, NodeProcessor>();
opRules.put(new RuleRegExp("R1",
ReduceSinkOperator.getOperatorName() + "%" +
".*" +
LimitOperator.getOperatorName() + "%"),
new TopNReducer());
LimitPushdownContext context = new LimitPushdownContext(pctx.getConf());
Dispatcher disp = new DefaultRuleDispatcher(null, opRules, context);
GraphWalker ogw = new DefaultGraphWalker(disp);
List<Node> topNodes = new ArrayList<Node>(pctx.getTopOps().values());
ogw.startWalking(topNodes, null);
return pctx;
}
private static class TopNReducer implements NodeProcessor {
public Object process(Node nd, Stack<Node> stack,
NodeProcessorCtx procCtx, Object... nodeOutputs) throws SemanticException {
ReduceSinkOperator rs = null;
for (int i = stack.size() - 2 ; i >= 0; i--) {
Operator<?> operator = (Operator<?>) stack.get(i);
if (operator.getNumChild() != 1) {
return false; // multi-GBY single-RS (TODO)
}
if (operator instanceof ReduceSinkOperator) {
rs = (ReduceSinkOperator) operator;
break;
}
if (!operator.acceptLimitPushdown()) {
return false;
}
}
if (rs != null) {
List<List<Integer>> distincts = rs.getConf().getDistinctColumnIndices();
if (distincts != null && distincts.size() > 1) {
// multi distinct case. can not sure that it's safe just by multiplying limit value
return false;
}
LimitOperator limit = (LimitOperator) nd;
rs.getConf().setTopN(limit.getConf().getLimit());
rs.getConf().setTopNMemoryUsage(((LimitPushdownContext) procCtx).threshold);
if (rs.getNumChild() == 1 && rs.getChildren().get(0) instanceof GroupByOperator) {
rs.getConf().setMapGroupBy(true);
}
}
return true;
}
}
private static class LimitPushdownContext implements NodeProcessorCtx {
private float threshold;
public LimitPushdownContext(HiveConf conf) throws SemanticException {
threshold = conf.getFloatVar(HiveConf.ConfVars.HIVELIMITPUSHDOWNMEMORYUSAGE);
if (threshold <= 0 || threshold >= 1) {
throw new SemanticException("Invalid memory usage value " + threshold +
" for " + HiveConf.ConfVars.HIVELIMITPUSHDOWNMEMORYUSAGE);
}
}
}
}