-
Notifications
You must be signed in to change notification settings - Fork 1.2k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
optimize left join predicate and eliminate useless appendVertices (#4980
) * optimize left join predicate & Eliminate useless appendVertices * fix * update * address leftExprs&rightExprs * address comments
- Loading branch information
Showing
7 changed files
with
295 additions
and
14 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
167 changes: 167 additions & 0 deletions
167
src/graph/optimizer/rule/OptimizeLeftJoinPredicateRule.cpp
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,167 @@ | ||
| // Copyright (c) 2022 vesoft inc. All rights reserved. | ||
| // | ||
| // This source code is licensed under Apache 2.0 License. | ||
|
|
||
| #include "graph/optimizer/rule/OptimizeLeftJoinPredicateRule.h" | ||
|
|
||
| #include "graph/optimizer/OptContext.h" | ||
| #include "graph/optimizer/OptGroup.h" | ||
| #include "graph/planner/plan/PlanNode.h" | ||
| #include "graph/planner/plan/Query.h" | ||
| #include "graph/util/ExpressionUtils.h" | ||
|
|
||
| using nebula::graph::PlanNode; | ||
| using nebula::graph::QueryContext; | ||
|
|
||
| namespace nebula { | ||
| namespace opt { | ||
|
|
||
| std::unique_ptr<OptRule> OptimizeLeftJoinPredicateRule::kInstance = | ||
| std::unique_ptr<OptimizeLeftJoinPredicateRule>(new OptimizeLeftJoinPredicateRule()); | ||
|
|
||
| OptimizeLeftJoinPredicateRule::OptimizeLeftJoinPredicateRule() { | ||
| RuleSet::QueryRules().addRule(this); | ||
| } | ||
|
|
||
| const Pattern& OptimizeLeftJoinPredicateRule::pattern() const { | ||
| static Pattern pattern = Pattern::create( | ||
| PlanNode::Kind::kHashLeftJoin, | ||
| {Pattern::create(PlanNode::Kind::kUnknown), | ||
| Pattern::create(PlanNode::Kind::kProject, | ||
| {Pattern::create(PlanNode::Kind::kAppendVertices, | ||
| {Pattern::create(PlanNode::Kind::kTraverse)})})}); | ||
| return pattern; | ||
| } | ||
|
|
||
| StatusOr<OptRule::TransformResult> OptimizeLeftJoinPredicateRule::transform( | ||
| OptContext* octx, const MatchedResult& matched) const { | ||
| auto* leftJoinGroupNode = matched.node; | ||
| auto* leftJoinGroup = leftJoinGroupNode->group(); | ||
| auto* leftJoin = static_cast<graph::HashLeftJoin*>(leftJoinGroupNode->node()); | ||
|
|
||
| auto* projectGroupNode = matched.dependencies[1].node; | ||
| auto* project = static_cast<graph::Project*>(projectGroupNode->node()); | ||
|
|
||
| auto* appendVerticesGroupNode = matched.dependencies[1].dependencies[0].node; | ||
| auto appendVertices = | ||
| static_cast<graph::AppendVertices*>(matched.dependencies[1].dependencies[0].node->node()); | ||
|
|
||
| auto traverse = static_cast<graph::Traverse*>( | ||
| matched.dependencies[1].dependencies[0].dependencies[0].node->node()); | ||
|
|
||
| auto& avNodeAlias = appendVertices->nodeAlias(); | ||
|
|
||
| auto& tvEdgeAlias = traverse->edgeAlias(); | ||
|
|
||
| auto& leftExprs = leftJoin->hashKeys(); | ||
| auto& rightExprs = leftJoin->probeKeys(); | ||
|
|
||
| bool found = false; | ||
| size_t rightExprIdx = 0; | ||
| for (size_t i = 0; i < rightExprs.size(); ++i) { | ||
| auto* rightExpr = rightExprs[i]; | ||
| if (rightExpr->kind() != Expression::Kind::kFunctionCall) { | ||
| continue; | ||
| } | ||
| auto* func = static_cast<FunctionCallExpression*>(rightExpr); | ||
| if (func->name() != "id" && func->name() != "_joinkey") { | ||
| continue; | ||
| } | ||
| auto& args = func->args()->args(); | ||
| DCHECK_EQ(args.size(), 1); | ||
| auto* arg = args[0]; | ||
| if (arg->kind() != Expression::Kind::kInputProperty) { | ||
| continue; | ||
| } | ||
| auto& alias = static_cast<InputPropertyExpression*>(arg)->prop(); | ||
| if (alias != avNodeAlias) continue; | ||
| // Must check if left exprs contain the same key | ||
| if (*leftExprs[i] != *rightExpr) { | ||
| return TransformResult::noTransform(); | ||
| } | ||
| if (found) { | ||
| return TransformResult::noTransform(); | ||
| } | ||
| rightExprIdx = i; | ||
| found = true; | ||
| } | ||
| if (!found) { | ||
| return TransformResult::noTransform(); | ||
| } | ||
|
|
||
| found = false; | ||
| size_t prjIdx = 0; | ||
| auto* columns = project->columns(); | ||
| for (size_t i = 0; i < columns->size(); ++i) { | ||
| const auto* col = columns->columns()[i]; | ||
| if (col->expr()->kind() != Expression::Kind::kInputProperty) { | ||
| continue; | ||
| } | ||
| auto* inputProp = static_cast<InputPropertyExpression*>(col->expr()); | ||
| if (inputProp->prop() != avNodeAlias) continue; | ||
| if (found) { | ||
| return TransformResult::noTransform(); | ||
| } | ||
| prjIdx = i; | ||
| found = true; | ||
| } | ||
| if (!found) { | ||
| return TransformResult::noTransform(); | ||
| } | ||
|
|
||
| auto* pool = octx->qctx()->objPool(); | ||
| // Let the new project generate expr `none_direct_dst($-.tvEdgeAlias)`, | ||
| // and let the new left join use it as right expr | ||
| auto* args = ArgumentList::make(pool); | ||
| args->addArgument(InputPropertyExpression::make(pool, tvEdgeAlias)); | ||
| auto* newPrjExpr = FunctionCallExpression::make(pool, "none_direct_dst", args); | ||
|
|
||
| auto oldYieldColumns = project->columns()->columns(); | ||
| auto* newYieldColumns = pool->makeAndAdd<YieldColumns>(); | ||
| for (size_t i = 0; i < oldYieldColumns.size(); ++i) { | ||
| if (i == prjIdx) { | ||
| newYieldColumns->addColumn(new YieldColumn(newPrjExpr, avNodeAlias)); | ||
| } else { | ||
| newYieldColumns->addColumn(oldYieldColumns[i]->clone().release()); | ||
| } | ||
| } | ||
| auto* newProject = graph::Project::make(octx->qctx(), nullptr, newYieldColumns); | ||
|
|
||
| // $-.`avNodeAlias` | ||
| auto* newRightExpr = InputPropertyExpression::make(pool, avNodeAlias); | ||
| std::vector<Expression*> newRightExprs; | ||
| for (size_t i = 0; i < rightExprs.size(); ++i) { | ||
| if (i == rightExprIdx) { | ||
| newRightExprs.emplace_back(newRightExpr); | ||
| } else { | ||
| newRightExprs.emplace_back(rightExprs[i]->clone()); | ||
| } | ||
| } | ||
| auto* newLeftJoin = | ||
| graph::HashLeftJoin::make(octx->qctx(), nullptr, nullptr, leftExprs, newRightExprs); | ||
|
|
||
| TransformResult result; | ||
| result.eraseAll = true; | ||
|
|
||
| newProject->setInputVar(appendVertices->inputVar()); | ||
| auto newProjectGroup = OptGroup::create(octx); | ||
| auto* newProjectGroupNode = newProjectGroup->makeGroupNode(newProject); | ||
| newProjectGroupNode->setDeps(appendVerticesGroupNode->dependencies()); | ||
|
|
||
| newLeftJoin->setLeftVar(leftJoin->leftInputVar()); | ||
| newLeftJoin->setRightVar(newProject->outputVar()); | ||
| newLeftJoin->setOutputVar(leftJoin->outputVar()); | ||
| auto* newLeftJoinGroupNode = OptGroupNode::create(octx, newLeftJoin, leftJoinGroup); | ||
| newLeftJoinGroupNode->dependsOn(leftJoinGroupNode->dependencies()[0]); | ||
| newLeftJoinGroupNode->dependsOn(newProjectGroup); | ||
|
|
||
| result.newGroupNodes.emplace_back(newLeftJoinGroupNode); | ||
| return result; | ||
| } | ||
|
|
||
| std::string OptimizeLeftJoinPredicateRule::toString() const { | ||
| return "OptimizeLeftJoinPredicateRule"; | ||
| } | ||
|
|
||
| } // namespace opt | ||
| } // namespace nebula |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,46 @@ | ||
| /* Copyright (c) 2022 vesoft inc. All rights reserved. | ||
| * | ||
| * This source code is licensed under Apache 2.0 License. | ||
| */ | ||
|
|
||
| #pragma once | ||
|
|
||
| #include "graph/optimizer/OptRule.h" | ||
|
|
||
| namespace nebula { | ||
| namespace opt { | ||
| // Before: | ||
| // HashLeftJoin({id(v)}, {id(v)}) | ||
| // | | | ||
| // ... Project | ||
| // | | | ||
| // AppendVertices(v) AppendVertices(v) | ||
| // | | | ||
| // ... Traverse(e) | ||
| // | ||
| // After: | ||
| // HashLeftJoin({id(v)}, {$-.v}) | ||
| // | | | ||
| // ... Project(..., none_direct_dst(e) AS v) | ||
| // | | | ||
| // AppendVertices(v) Traverse(e) | ||
| // | | ||
| // ... | ||
| // | ||
| class OptimizeLeftJoinPredicateRule final : public OptRule { | ||
| public: | ||
| const Pattern &pattern() const override; | ||
|
|
||
| StatusOr<OptRule::TransformResult> transform(OptContext *qctx, | ||
| const MatchedResult &matched) const override; | ||
|
|
||
| std::string toString() const override; | ||
|
|
||
| private: | ||
| OptimizeLeftJoinPredicateRule(); | ||
|
|
||
| static std::unique_ptr<OptRule> kInstance; | ||
| }; | ||
|
|
||
| } // namespace opt | ||
| } // namespace nebula |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
64 changes: 64 additions & 0 deletions
64
tests/tck/features/optimizer/OptimizeLeftJoinPredicateRule.feature
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,64 @@ | ||
| # Copyright (c) 2022 vesoft inc. All rights reserved. | ||
| # | ||
| # This source code is licensed under Apache 2.0 License. | ||
| Feature: Optimize left join predicate | ||
|
|
||
| Background: | ||
| Given a graph with space named "nba" | ||
|
|
||
| Scenario: optimize left join predicate | ||
| When profiling query: | ||
| """ | ||
| MATCH (person:player)-[:like*1..2]-(friend:player)-[:serve]->(friendTeam:team) | ||
| WHERE id(person) == "Tony Parker" AND id(friend) != "Tony Parker" | ||
| WITH DISTINCT friend, friendTeam | ||
| OPTIONAL MATCH (friend)<-[:like]-(friend2:player)<-[:like]-(friendTeam) | ||
| WITH friendTeam, count(friend2) AS numFriends | ||
| RETURN | ||
| id(friendTeam) AS teamId, | ||
| friendTeam.team.name AS teamName, | ||
| numFriends | ||
| ORDER BY teamName DESC | ||
| """ | ||
| Then the result should be, in order, with relax comparison: | ||
| | teamId | teamName | numFriends | | ||
| | "Warriors" | "Warriors" | 0 | | ||
| | "Trail Blazers" | "Trail Blazers" | 0 | | ||
| | "Thunders" | "Thunders" | 0 | | ||
| | "Suns" | "Suns" | 0 | | ||
| | "Spurs" | "Spurs" | 0 | | ||
| | "Rockets" | "Rockets" | 0 | | ||
| | "Raptors" | "Raptors" | 0 | | ||
| | "Pistons" | "Pistons" | 0 | | ||
| | "Magic" | "Magic" | 0 | | ||
| | "Lakers" | "Lakers" | 0 | | ||
| | "Kings" | "Kings" | 0 | | ||
| | "Jazz" | "Jazz" | 0 | | ||
| | "Hornets" | "Hornets" | 0 | | ||
| | "Heat" | "Heat" | 0 | | ||
| | "Hawks" | "Hawks" | 0 | | ||
| | "Grizzlies" | "Grizzlies" | 0 | | ||
| | "Clippers" | "Clippers" | 0 | | ||
| | "Celtics" | "Celtics" | 0 | | ||
| | "Cavaliers" | "Cavaliers" | 0 | | ||
| | "Bulls" | "Bulls" | 0 | | ||
| | "76ers" | "76ers" | 0 | | ||
| And the execution plan should be: | ||
| | id | name | dependencies | operator info | | ||
| | 21 | Sort | 18 | | | ||
| | 18 | Project | 17 | | | ||
| | 17 | Aggregate | 16 | | | ||
| | 16 | HashLeftJoin | 10,15 | {"hashKeys": ["_joinkey($-.friendTeam)", "_joinkey($-.friend)"], "probeKeys": ["$-.friendTeam", "_joinkey($-.friend)"]} | | ||
| | 10 | Dedup | 28 | | | ||
| | 28 | Project | 22 | | | ||
| | 22 | Filter | 26 | | | ||
| | 26 | AppendVertices | 25 | | | ||
| | 25 | Traverse | 24 | | | ||
| | 24 | Traverse | 2 | | | ||
| | 2 | Dedup | 1 | | | ||
| | 1 | PassThrough | 3 | | | ||
| | 3 | Start | | | | ||
| | 15 | Project | 14 | {"columns": ["$-.friend AS friend", "$-.friend2 AS friend2", "none_direct_dst($-.__VAR_3) AS friendTeam"]} | | ||
| | 14 | Traverse | 12 | | | ||
| | 12 | Traverse | 11 | | | ||
| | 11 | Argument | | | |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters