Add Exchange before GroupId to improve Partial Aggregation#11741
Closed
findepi wants to merge 2 commits intoprestodb:masterfrom
findepi:findepi/master/add-exchange-before-groupid-to-improve-partial-aggregation-16b5ea
Closed
Add Exchange before GroupId to improve Partial Aggregation#11741findepi wants to merge 2 commits intoprestodb:masterfrom findepi:findepi/master/add-exchange-before-groupid-to-improve-partial-aggregation-16b5ea
findepi wants to merge 2 commits intoprestodb:masterfrom
findepi:findepi/master/add-exchange-before-groupid-to-improve-partial-aggregation-16b5ea
Conversation
The idea was abandoned during #11267 review.
The rule brings significant improvement in TPC-DS Q22 and Q67 while not causing much regression in other TPC-H, TPC-DS queries. (The only observably regressing queries were still much better than non-CBO baseline.)
kokosing
approved these changes
Oct 29, 2018
Contributor
kokosing
left a comment
kokosing
left a comment
There was a problem hiding this comment.
LGTM because it already went through the internal review process.
Some nit comments.
| * GroupId (before multiplication) makes partial aggregation more effective, resulting in less data being | ||
| * exchanged afterwards. | ||
| */ | ||
| public class AddExchangesBelowPartialAggregationOverGroupIdRuleSet |
Contributor
There was a problem hiding this comment.
It would be nice to have some unit tests, but that would require a lot of plumbing.
| return transform(aggregation, groupId, context) | ||
| .map(newAggregation -> { | ||
| PlanNode newExchange = exchange.replaceChildren(ImmutableList.of(newAggregation)); | ||
| return Result.ofPlanNode(newExchange); |
|
|
||
| List<Symbol> desiredHashSymbols = groupingSetHistogram.entrySet().stream() | ||
| // Take only frequently used symbols | ||
| .filter(entry -> entry.getCount() >= groupId.getGroupingSets().size() * GROUPING_SETS_SYMBOL_REQUIRED_FREQUENCY) |
Contributor
There was a problem hiding this comment.
what if all symbols are less frequent, shouldn't we return Optional.empty() here (short circuit)?
|
|
||
| StreamPreferredProperties requiredProperties = fixedParallelism().withPartitioning(desiredHashSymbols); | ||
| StreamProperties sourceProperties = derivePropertiesRecursively(groupId.getSource(), context); | ||
| if (requiredProperties.isSatisfiedBy(sourceProperties)) { |
Contributor
There was a problem hiding this comment.
it would be nice to extract this as a separate rule that would remove unnecessary partial aggregations
arhimondr
reviewed
Nov 16, 2018
Member
arhimondr
left a comment
arhimondr
left a comment
There was a problem hiding this comment.
Could you please provide some high level overview for the problem you are trying to solve?
| public static final String LEGACY_ROW_FIELD_ORDINAL_ACCESS = "legacy_row_field_ordinal_access"; | ||
| public static final String ITERATIVE_OPTIMIZER = "iterative_optimizer_enabled"; | ||
| public static final String ITERATIVE_OPTIMIZER_TIMEOUT = "iterative_optimizer_timeout"; | ||
| public static final String ENABLE_FORCED_EXCHANGE_BELOW_GROUP_ID = "enable_forced_exchange_below_group_id"; |
Member
There was a problem hiding this comment.
Maybe add_exchange_before_group_id
| private double spillMaxUsedSpaceThreshold = 0.9; | ||
| private boolean iterativeOptimizerEnabled = true; | ||
| private boolean enableStatsCalculator = true; | ||
| private boolean enableForcedExchangeBelowGroupId = true; |
Member
There was a problem hiding this comment.
Can we have it disabled by default?
| new PushPartialAggregationThroughJoin(), | ||
| new PushPartialAggregationThroughExchange(metadata.getFunctionRegistry()), | ||
| new PruneJoinColumns()))); | ||
| builder.add(new IterativeOptimizer( |
Member
There was a problem hiding this comment.
This rule uses cost model. Cost model loses estimates when there are any partial aggregation in between.
| * GroupId (before multiplication) makes partial aggregation more effective, resulting in less data being | ||
| * exchanged afterwards. | ||
| */ | ||
| public class AddExchangesBelowPartialAggregationOverGroupIdRuleSet |
Member
There was a problem hiding this comment.
Maybe just AddExchangesBeforeGroupId?
findepi
deleted the
findepi/master/add-exchange-before-groupid-to-improve-partial-aggregation-16b5ea
branch
6 tasks
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The rule brings significant improvement in TPC-DS Q22 and Q67 while not
causing much regression in other TPC-H, TPC-DS queries. (The only
observably regressing queries were still much better than non-CBO
baseline.)