github.com/matrixorigin/matrixone@v1.2.0/pkg/sql/plan/associative_law.go (about)

     1  // Copyright 2022 Matrix Origin
     2  //
     3  // Licensed under the Apache License, Version 2.0 (the "License");
     4  // you may not use this file except in compliance with the License.
     5  // You may obtain a copy of the License at
     6  //
     7  //      http://www.apache.org/licenses/LICENSE-2.0
     8  //
     9  // Unless required by applicable law or agreed to in writing, software
    10  // distributed under the License is distributed on an "AS IS" BASIS,
    11  // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
    12  // See the License for the specific language governing permissions and
    13  // limitations under the License.
    14  
    15  package plan
    16  
    17  import "github.com/matrixorigin/matrixone/pkg/pb/plan"
    18  
    19  // for A*(B*C), if C.sel>0.9 and B<C, change this to (A*B)*C
    20  func (builder *QueryBuilder) applyAssociativeLawRule1(nodeID int32) int32 {
    21  	node := builder.qry.Nodes[nodeID]
    22  	if len(node.Children) > 0 {
    23  		for i, child := range node.Children {
    24  			node.Children[i] = builder.applyAssociativeLawRule1(child)
    25  		}
    26  	}
    27  	if node.NodeType != plan.Node_JOIN || node.JoinType != plan.Node_INNER {
    28  		return nodeID
    29  	}
    30  	rightChild := builder.qry.Nodes[node.Children[1]]
    31  	if rightChild.NodeType != plan.Node_JOIN || rightChild.JoinType != plan.Node_INNER {
    32  		return nodeID
    33  	}
    34  	NodeB := builder.qry.Nodes[rightChild.Children[0]]
    35  	NodeC := builder.qry.Nodes[rightChild.Children[1]]
    36  	if NodeC.Stats.Selectivity < 0.9 || NodeB.Stats.Outcnt >= NodeC.Stats.Outcnt {
    37  		return nodeID
    38  	}
    39  	node.Children[1] = NodeB.NodeId
    40  	determineHashOnPK(node.NodeId, builder)
    41  	if !node.Stats.HashmapStats.HashOnPK {
    42  		// a join b must be hash on primary key, or we can not do this change
    43  		node.Children[1] = rightChild.NodeId
    44  		return node.NodeId
    45  	}
    46  	rightChild.Children[0] = node.NodeId
    47  	ReCalcNodeStats(rightChild.NodeId, builder, true, false, true)
    48  	return rightChild.NodeId
    49  }
    50  
    51  // for （A*B)*C, if C.sel<0.5, change this to A*(B*C)
    52  func (builder *QueryBuilder) applyAssociativeLawRule2(nodeID int32) int32 {
    53  	node := builder.qry.Nodes[nodeID]
    54  	if len(node.Children) > 0 {
    55  		for i, child := range node.Children {
    56  			node.Children[i] = builder.applyAssociativeLawRule2(child)
    57  		}
    58  	}
    59  	if node.NodeType != plan.Node_JOIN || node.JoinType != plan.Node_INNER {
    60  		return nodeID
    61  	}
    62  	leftChild := builder.qry.Nodes[node.Children[0]]
    63  	if leftChild.NodeType != plan.Node_JOIN || leftChild.JoinType != plan.Node_INNER {
    64  		return nodeID
    65  	}
    66  	NodeC := builder.qry.Nodes[node.Children[1]]
    67  	if NodeC.Stats.Selectivity > 0.5 {
    68  		return nodeID
    69  	}
    70  	NodeB := builder.qry.Nodes[leftChild.Children[1]]
    71  	node.Children[0] = NodeB.NodeId
    72  	determineHashOnPK(node.NodeId, builder)
    73  	if !node.Stats.HashmapStats.HashOnPK {
    74  		// b join c must be hash on primary key, or we can not do this change
    75  		node.Children[0] = leftChild.NodeId
    76  		return node.NodeId
    77  	}
    78  	leftChild.Children[1] = node.NodeId
    79  	ReCalcNodeStats(leftChild.NodeId, builder, true, false, true)
    80  	return leftChild.NodeId
    81  }
    82  
    83  // for （A*B)*C, if A.outcnt>C.outcnt and C.sel<B.sel, change this to (A*C)*B
    84  func (builder *QueryBuilder) applyAssociativeLawRule3(nodeID int32) int32 {
    85  	node := builder.qry.Nodes[nodeID]
    86  	if len(node.Children) > 0 {
    87  		for i, child := range node.Children {
    88  			node.Children[i] = builder.applyAssociativeLawRule3(child)
    89  		}
    90  	}
    91  	if node.NodeType != plan.Node_JOIN || node.JoinType != plan.Node_INNER {
    92  		return nodeID
    93  	}
    94  	leftChild := builder.qry.Nodes[node.Children[0]]
    95  	if leftChild.NodeType != plan.Node_JOIN || leftChild.JoinType != plan.Node_INNER {
    96  		return nodeID
    97  	}
    98  	NodeA := builder.qry.Nodes[leftChild.Children[0]]
    99  	NodeB := builder.qry.Nodes[leftChild.Children[1]]
   100  	NodeC := builder.qry.Nodes[node.Children[1]]
   101  	if NodeA.Stats.Outcnt < NodeC.Stats.Outcnt || NodeC.Stats.Selectivity >= NodeB.Stats.Selectivity {
   102  		return nodeID
   103  	}
   104  
   105  	node.Children[0] = NodeA.NodeId
   106  	newSel := NodeC.Stats.Selectivity / getHashColsNDVRatio(node.NodeId, builder)
   107  	if newSel >= NodeB.Stats.Selectivity {
   108  		//new selectivity bigger than b.sel, can't do this change
   109  		node.Children[0] = leftChild.NodeId
   110  		return node.NodeId
   111  	}
   112  	leftChild.Children[0] = node.NodeId
   113  	ReCalcNodeStats(leftChild.NodeId, builder, true, false, true)
   114  	return leftChild.NodeId
   115  }
   116  
   117  func (builder *QueryBuilder) applyAssociativeLaw(nodeID int32) int32 {
   118  	if builder.optimizerHints != nil && builder.optimizerHints.joinOrdering != 0 {
   119  		return nodeID
   120  	}
   121  	nodeID = builder.applyAssociativeLawRule1(nodeID)
   122  	builder.determineBuildAndProbeSide(nodeID, true)
   123  	nodeID = builder.applyAssociativeLawRule2(nodeID)
   124  	builder.determineBuildAndProbeSide(nodeID, true)
   125  	nodeID = builder.applyAssociativeLawRule3(nodeID)
   126  	builder.determineBuildAndProbeSide(nodeID, true)
   127  	return nodeID
   128  }