[enhancement](nereids) improve lots of values in insert into values statement (#40202) (#51925) #51949
Conversation
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
… statement (apache#40202) improve lots of values in `insert into values` statement by bypass NereidsPlanner the main logic is 1. `InsertUtils.normalizePlan` use `FoldConstantRuleOnFE` to reduce the expression, e.g. `values(date(now())` 2. `FastInsertIntoValuesPlanner` skip most of rules to analyze and rewrite `LogicalInlineTable` to `LogicalUnion` or `LogicalOneRowRelation` 3. fast parse date time string without date format 4. getHintMap and normal lexer share the same tokens 5. `set enable_fast_analyze_into_values=false` can force to execute all optimize rules, when we meet some bugs in `FastInsertIntoValuesPlanner` test: insert 1000 rows with 1000 columns, the columns contains int, bigint, decimal(26,7), date, datetime, varchar(10 chinese chars) +---------------------------------+------------------------------------------------------+--------------------------+--------------------------+ |FastInsertIntoValuesPlanner |NereidsPlanner(enable_fast_analyze_into_values=false) |Legacy optimizer in 2.1.6 | Nereids planner in 2.1.6 | +---------------------------------+------------------------------------------------------+--------------------------+--------------------------+ |16s(bottleneck is antlr's lexer) |32s |16s |80s | +---------------------------------+------------------------------------------------------+--------------------------+--------------------------+ If you use FastInsertIntoValuesPlanner with group commit in a transaction, the time can reduce to 12s. TODO: build a custom lexer. in my hand write lexer test, FastInsertIntoValuesPlanner without group commit can reduce 16s to 12s, but it will take more effort: RegularExpression -> NFA -> DFA -> minimal DFA -> Lexer codegen (cherry picked from commit 81f3c48)
Contributor
|
Thank you for your contribution to Apache Doris. Please clearly describe your PR:
|
Contributor
Author
|
run buildall |
Contributor
Author
|
run buildall |
Contributor
Author
|
run buildall |
924060929
changed the title
TPC-H: Total hot run time: 39974 ms |
TPC-DS: Total hot run time: 190000 ms |
ClickBench: Total hot run time: 30.33 s |
morrySnow
approved these changes
Jun 20, 2025
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.
cherry pick from #40202 and #51925