[fix] (planner) support auto aggregation for random distributed table #33630
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
|
Thank you for your contribution to Apache Doris. Since 2024-03-18, the Document has been moved to doris-website. |
Contributor
Author
|
run buildall |
TPC-H: Total hot run time: 38165 ms |
TPC-DS: Total hot run time: 184601 ms |
ClickBench: Total hot run time: 30.2 s |
|
Load test result on machine: 'aliyun_ecs.c7a.8xlarge_32C64G' |
Contributor
Author
|
run buildall |
TPC-H: Total hot run time: 38061 ms |
TPC-DS: Total hot run time: 183526 ms |
ClickBench: Total hot run time: 30.35 s |
|
Load test result on machine: 'aliyun_ecs.c7a.8xlarge_32C64G' |
morrySnow
reviewed
Apr 15, 2024
fe/fe-core/src/main/java/org/apache/doris/analysis/StmtRewriter.java
Outdated
Show resolved
Hide resolved
DarvenDuan
force-pushed
the
optimize_random_distribute
branch
from
26c6eeb to
e005f3e
Compare
DarvenDuan
force-pushed
the
optimize_random_distribute
branch
4 times, most recently
from
c932047 to
d2150a8
Compare
Contributor
Author
|
run p0 |
Contributor
Author
|
run buildall |
TPC-H: Total hot run time: 38329 ms |
TPC-DS: Total hot run time: 186263 ms |
ClickBench: Total hot run time: 30.35 s |
Contributor
Author
|
run p0 |
|
Load test result on machine: 'aliyun_ecs.c7a.8xlarge_32C64G' |
| public List<Rule> buildRules() { | ||
| return ImmutableList.of( | ||
| // Project(Scan) | ||
| logicalProject(logicalOlapScan()).when(project -> isRandomDistributedTbl(project.child())) |
Contributor
There was a problem hiding this comment.
better to simply this rule by only match logicalOlapScan with randomDist, then always gererate a agg node above it. The MergeAggregate rule can merge redundant agg later.
Contributor
Author
There was a problem hiding this comment.
check agg(olapScan) there will avoid to rewrite olapScan to agg(olapScan) twice.
| if (col.isAggregated()) { | ||
| AggregateType aggType = col.getAggregationType(); | ||
| // agg type not mach | ||
| return (aggType == AggregateType.SUM || aggType == AggregateType.MAX |
Contributor
There was a problem hiding this comment.
we have to deal with other agg types, including HLL_UNION, BITMAP_UNION, QUANTILE_UNION and GENERIC(agg_state related)
Contributor
|
Generally, we can only fix the bug for nereids planner, unless there is a strong needs for branch 2.0.(the old planner will be discarded very soon). And we should consider not only MAX, MIN, SUM and also other agg types like HLL_UNION, BITMAP_UNION, QUANTILE_UNION and GENERIC(agg_state related) |
DarvenDuan
force-pushed
the
optimize_random_distribute
branch
from
da53339 to
3d41b98
Compare
DarvenDuan
force-pushed
the
optimize_random_distribute
branch
from
3d41b98 to
d4a168d
Compare
Contributor
|
run buildall |
Contributor
Author
|
run p0 |
github-actions
bot
added
the
approved
Contributor
|
PR approved by at least one committer and no changes requested. |
morrySnow
approved these changes
May 17, 2024
yiguolei
pushed a commit
that referenced
this pull request
May 18, 2024
…ed table (#33630) support auto aggregation for querying detail data of random distributed table: the same key column will return only one row.
dataroaring
pushed a commit
that referenced
this pull request
May 26, 2024
…ed table (#33630) support auto aggregation for querying detail data of random distributed table: the same key column will return only one row.
dataroaring
pushed a commit
that referenced
this pull request
Jul 24, 2024
## Proposed changes Bug description: In PR #33630, Doris supports auto aggregation for random distributed table, but it not only effects query statements, so if we delete from a random distributed table, will get an error because of unexpectedly rewriting. ``` CREATE TABLE `test_tbl` ( `k` INT NULL, `v` BIGINT SUM NULL ) ENGINE=OLAP AGGREGATE KEY(`k`) DISTRIBUTED BY RANDOM BUCKETS AUTO; mysql > delete from test_tbl where k=1; ERROR 1105 (HY000): errCode = 2, detailMessage = Where clause only supports compound predicate, binary predicate, is_null predicate or in predicate. ``` fix: Check whether it is a query statement before rewriting.
dataroaring
pushed a commit
that referenced
this pull request
Jul 24, 2024
## Proposed changes Bug description: In PR #33630, Doris supports auto aggregation for random distributed table, but it not only effects query statements, so if we delete from a random distributed table, will get an error because of unexpectedly rewriting. ``` CREATE TABLE `test_tbl` ( `k` INT NULL, `v` BIGINT SUM NULL ) ENGINE=OLAP AGGREGATE KEY(`k`) DISTRIBUTED BY RANDOM BUCKETS AUTO; mysql > delete from test_tbl where k=1; ERROR 1105 (HY000): errCode = 2, detailMessage = Where clause only supports compound predicate, binary predicate, is_null predicate or in predicate. ``` fix: Check whether it is a query statement before rewriting.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 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.
Proposed changes
Issue Number: close #xxx
support auto aggregation for querying detail data of random distributed table: the same key column will return only one row.
step 1: fix legacy planner (√)
step 2: fix nereids planner (√)
Further comments
If this is a relatively large or complex change, kick off the discussion at dev@doris.apache.org by explaining why you chose the solution you did and what alternatives you considered, etc...