*: global index support `index_merge` and `mem_index_merge` #52971

Defined2014 · 2024-04-29T06:32:24Z

What problem does this PR solve?

Issue Number: close #43013

Problem Summary:

What changed and how does it work?

Check List

Tests

Unit test
Integration test
Manual test (add detailed scripts or steps below)
No need to test
- I checked and no code files have been changed.

Side effects

Performance regression: Consumes more CPU
Performance regression: Consumes more Memory
Breaking backward compatibility

Documentation

Release note

Please refer to Release Notes Language Style Guide to write a quality release note.

None

tiprow · 2024-04-29T06:32:53Z

Hi @Defined2014. Thanks for your PR.

PRs from untrusted users cannot be marked as trusted with /ok-to-test in this repo meaning untrusted PR authors can never trigger tests themselves. Collaborators can still trigger tests on the PR using /test all.

I understand the commands that are listed here.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

codecov · 2024-04-29T07:01:35Z

Codecov Report

Attention: Patch coverage is 89.47368% with 14 lines in your changes are missing coverage. Please review.

Project coverage is 74.0324%. Comparing base (3fc5704) to head (743b8ec).
Report is 25 commits behind head on master.

Additional details and impacted files

@@               Coverage Diff                @@
##             master     #52971        +/-   ##
================================================
+ Coverage   72.3804%   74.0324%   +1.6520%     
================================================
  Files          1482       1508        +26     
  Lines        428743     440608     +11865     
================================================
+ Hits         310326     326193     +15867     
+ Misses        99117      94208      -4909     
- Partials      19300      20207       +907

Flag	Coverage Δ
integration	`48.8144% <88.7218%> (?)`
unit	`71.2354% <63.9097%> (+0.0176%)`	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
dumpling	`53.9957% <ø> (ø)`
parser	`∅ <ø> (∅)`
br	`44.2196% <ø> (+3.0757%)`	⬆️

Defined2014 · 2024-04-29T08:16:40Z

/retest

tiprow · 2024-04-29T08:18:43Z

@Defined2014: Cannot trigger testing until a trusted user reviews the PR and leaves an /ok-to-test message.

In response to this:

/retest

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

mjonss

LGTM, with some minor suggestions and questions.

pkg/executor/index_merge_reader.go

pkg/executor/mem_reader.go

mjonss · 2024-04-29T11:05:53Z

pkg/executor/mem_reader.go

+					cnt := 1
+					hMap.Set(handle, &cnt)
+				} else {
+					*(v.(*int))++


I don't know much about this part of the executor, can you help me understand what this means? That only if every memory reader has a match, it should be included if intersect later?
Can it be that a single (reader) handle is found multiple times for the same reader?

That only if every memory reader has a match, it should be included if intersect later?

Yes.

Can it be that a single (reader) handle is found multiple times for the same reader?

No, each handle will only appear once in the same index worker.

pkg/planner/core/find_best_task.go

pkg/planner/core/indexmerge_path.go

tests/integrationtest/t/globalindex/mem_index_merge.test

Defined2014 · 2024-04-30T01:57:05Z

/retest

tiprow · 2024-04-30T01:58:27Z

@Defined2014: Cannot trigger testing until a trusted user reviews the PR and leaves an /ok-to-test message.

In response to this:

/retest

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

Defined2014 · 2024-04-30T04:46:07Z

/hold Awaiting review by planner team and executor team.

AilinKid

Rest LGTM

AilinKid · 2024-05-06T07:14:36Z

pkg/executor/mem_reader.go

+		// [partNum][rangeNum]
+		var readerKvRanges [][]kv.KeyRange
+		if m.partitionMode {
+			if r, ok := reader.(*memIndexReader); ok && r.index.Global {


seems it's part of range build-related logic, can we move this to buildTableKeyRanges?

and here we can only start with L1044

Done, PTAL @AilinKid

pkg/planner/core/indexmerge_path.go

AilinKid · 2024-05-06T07:35:56Z

pkg/executor/mem_reader.go

+			readerKvRanges = [][]kv.KeyRange{m.indexMergeReader.keyRanges[i]}
+		}
+
+		for j, kr := range readerKvRanges {


this function looks so long, why remove the original encapsulation here?

Because the original loop order is

for each partition { for each index { for each range { .... } } do union or intersection }

Now I changed it to

for each index { for each partition { for each range { .... } } } do union or intersection

The order of loops is same as indexMergeExec. The reason is for global index which is in partition mode, but it only has one partition. And it's hard to create table ranges by distsql.TableHandlesToKVRanges function.

Done, I divided this function into two functions

Defined2014 · 2024-05-06T10:19:28Z

/test pull-mysql-client-test

Defined2014 · 2024-05-06T10:59:55Z

/test check-dev2

ti-chi-bot · 2024-05-07T03:16:52Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: AilinKid, mjonss

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

~~OWNERS~~ [AilinKid,mjonss]
~~pkg/planner/OWNERS~~ [AilinKid]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

ti-chi-bot · 2024-05-07T03:16:55Z

[LGTM Timeline notifier]

Timeline:

2024-04-29 12:42:55.849555681 +0000 UTC m=+274729.606691255: ☑️ agreed by mjonss.
2024-05-07 03:16:54.648737871 +0000 UTC m=+931968.405873444: ☑️ agreed by AilinKid.

Defined2014 · 2024-05-07T03:22:15Z

/unhold

Defined2014 · 2024-05-07T03:54:08Z

/retest

Defined2014 · 2024-05-07T04:15:45Z

/test mysql-test

ti-chi-bot · 2024-05-07T04:15:47Z

@Defined2014: The specified target(s) for /test were not found.
The following commands are available to trigger required jobs:

/test build
/test check-dev
/test check-dev2
/test mysql-test
/test pull-br-integration-test
/test pull-integration-ddl-test
/test pull-lightning-integration-test
/test pull-mysql-client-test
/test unit-test

The following commands are available to trigger optional jobs:

/test canary-notify-when-compatibility-sections-changed
/test pingcap/tidb/canary_ghpr_unit_test
/test pull-common-test
/test pull-e2e-test
/test pull-integration-common-test
/test pull-integration-copr-test
/test pull-integration-jdbc-test
/test pull-integration-mysql-test
/test pull-integration-nodejs-test
/test pull-sqllogic-test
/test pull-tiflash-test

Use /test all to run the following jobs that were automatically triggered:

pingcap/tidb/ghpr_build
pingcap/tidb/ghpr_check
pingcap/tidb/ghpr_check2
pingcap/tidb/ghpr_mysql_test
pingcap/tidb/ghpr_unit_test
pingcap/tidb/pull_integration_ddl_test
pingcap/tidb/pull_mysql_client_test

In response to this:

/test mysql-test

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

Defined2014 · 2024-05-07T04:16:09Z

/test mysql-test

…52971) close pingcap#43013

global index support index_merge and mem_index_merge

6306251

ti-chi-bot bot added release-note-none size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files. sig/planner SIG: Planner labels Apr 29, 2024

Defined2014 added 2 commits April 29, 2024 14:39

update comments

c9bd518

update tests

e8a33ed

Defined2014 changed the title *: global index support index_merge and mem_index_merge *: global index support index_merge and mem_index_merge Apr 29, 2024

update tests

3537e1f

Defined2014 force-pushed the support-index-merge branch from 51a38ce to 3537e1f Compare April 29, 2024 07:46

update

41fa347

Defined2014 requested review from AilinKid, guo-shaoge, mjonss and tiancaiamao and removed request for AilinKid and guo-shaoge April 29, 2024 09:30

update comments

aa28e19

mjonss approved these changes Apr 29, 2024

View reviewed changes

ti-chi-bot bot added the needs-1-more-lgtm label Apr 29, 2024

follow comments

7009e62

ti-chi-bot bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Apr 30, 2024

AilinKid reviewed May 6, 2024

View reviewed changes

Defined2014 added 2 commits May 6, 2024 16:08

code clean

2258089

update

fac9084

Defined2014 force-pushed the support-index-merge branch from 3bf2bc8 to 2258089 Compare May 6, 2024 09:26

Defined2014 added 2 commits May 6, 2024 17:56

follow comments

fa25d13

reduce modify

5871f4c

split function

043b1a5

AilinKid approved these changes May 7, 2024

View reviewed changes

ti-chi-bot bot added approved lgtm and removed needs-1-more-lgtm labels May 7, 2024

ti-chi-bot bot removed the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label May 7, 2024

update

743b8ec

ti-chi-bot bot merged commit b66a85c into pingcap:master May 7, 2024
21 checks passed

Defined2014 deleted the support-index-merge branch May 7, 2024 04:44

3AceShowHand pushed a commit to 3AceShowHand/tidb that referenced this pull request May 7, 2024

*: global index support index_merge and mem_index_merge (pingcap#…

707258f

…52971) close pingcap#43013

terry1purcell pushed a commit to terry1purcell/tidb that referenced this pull request May 17, 2024

*: global index support index_merge and mem_index_merge (pingcap#…

10982e3

…52971) close pingcap#43013

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

*: global index support `index_merge` and `mem_index_merge` #52971

*: global index support `index_merge` and `mem_index_merge` #52971

Defined2014 commented Apr 29, 2024

tiprow bot commented Apr 29, 2024

codecov bot commented Apr 29, 2024 •

edited

Defined2014 commented Apr 29, 2024

tiprow bot commented Apr 29, 2024

mjonss left a comment

mjonss Apr 29, 2024

Defined2014 Apr 30, 2024 •

edited

Defined2014 commented Apr 30, 2024

tiprow bot commented Apr 30, 2024

Defined2014 commented Apr 30, 2024

AilinKid left a comment

AilinKid May 6, 2024

AilinKid May 6, 2024

Defined2014 May 6, 2024 •

edited

AilinKid May 6, 2024

Defined2014 May 6, 2024

Defined2014 May 7, 2024

Defined2014 commented May 6, 2024

Defined2014 commented May 6, 2024

ti-chi-bot bot commented May 7, 2024

ti-chi-bot bot commented May 7, 2024

Defined2014 commented May 7, 2024

Defined2014 commented May 7, 2024

Defined2014 commented May 7, 2024

ti-chi-bot bot commented May 7, 2024

Defined2014 commented May 7, 2024

*: global index support index_merge and mem_index_merge #52971

*: global index support index_merge and mem_index_merge #52971

Conversation

Defined2014 commented Apr 29, 2024

What problem does this PR solve?

What changed and how does it work?

Check List

Release note

tiprow bot commented Apr 29, 2024

codecov bot commented Apr 29, 2024 • edited

Codecov Report

Defined2014 commented Apr 29, 2024

tiprow bot commented Apr 29, 2024

mjonss left a comment

Choose a reason for hiding this comment

mjonss Apr 29, 2024

Choose a reason for hiding this comment

Defined2014 Apr 30, 2024 • edited

Choose a reason for hiding this comment

Defined2014 commented Apr 30, 2024

tiprow bot commented Apr 30, 2024

Defined2014 commented Apr 30, 2024

AilinKid left a comment

Choose a reason for hiding this comment

AilinKid May 6, 2024

Choose a reason for hiding this comment

AilinKid May 6, 2024

Choose a reason for hiding this comment

Defined2014 May 6, 2024 • edited

Choose a reason for hiding this comment

AilinKid May 6, 2024

Choose a reason for hiding this comment

Defined2014 May 6, 2024

Choose a reason for hiding this comment

Defined2014 May 7, 2024

Choose a reason for hiding this comment

Defined2014 commented May 6, 2024

Defined2014 commented May 6, 2024

ti-chi-bot bot commented May 7, 2024

ti-chi-bot bot commented May 7, 2024

[LGTM Timeline notifier]

Defined2014 commented May 7, 2024

Defined2014 commented May 7, 2024

Defined2014 commented May 7, 2024

ti-chi-bot bot commented May 7, 2024

Defined2014 commented May 7, 2024

*: global index support `index_merge` and `mem_index_merge` #52971

*: global index support `index_merge` and `mem_index_merge` #52971

codecov bot commented Apr 29, 2024 •

edited

Defined2014 Apr 30, 2024 •

edited

Defined2014 May 6, 2024 •

edited