Skip to content

[EXPERIMENTAL] VPlayer parallel applier#17370

Closed
shlomi-noach wants to merge 46 commits intovitessio:mainfrom
planetscale:vplayer-parallel-applier
Closed

[EXPERIMENTAL] VPlayer parallel applier#17370
shlomi-noach wants to merge 46 commits intovitessio:mainfrom
planetscale:vplayer-parallel-applier

Conversation

@shlomi-noach
Copy link
Copy Markdown
Contributor

@shlomi-noach shlomi-noach commented Dec 11, 2024

Description

EXPERIMENTAL only. Details to come. Right now just looking at a potential for parallel applier based on WRITESET and benchmarking parallel vs. linear performance (local benchmarking is inconclusive).

Related Issue(s)

TBD.

Checklist

  • "Backport to:" labels have been added if this change should be back-ported to release branches
  • If this change is to be back-ported to previous releases, a justification is included in the PR description
  • Tests were added or are not required
  • Did the new or modified tests pass consistently locally and on CI?
  • Documentation was added or is not required

Deployment Notes

@shlomi-noach
VReplication parallel applier; read sequence number and last committe…
c6f7c2e 
…d in GTID event (WRITESET)

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
last commited and sequence number must not change even if applying mu…
b8fefdf 
…ltiple events

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
generate CI for onlineddl_vrepl_bench
b2f0da8 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
cluster_endtoend_onlineddl_vrepl_bench.yml
72e0f9f 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
increase benchmark count and time
847fd52 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@vitess-bot
Copy link
Copy Markdown
Contributor

vitess-bot bot commented Dec 11, 2024

Review Checklist

Hello reviewers! 👋 Please follow this checklist when reviewing this Pull Request.

General

  • Ensure that the Pull Request has a descriptive title.
  • Ensure there is a link to an issue (except for internal cleanup and flaky test fixes), new features should have an RFC that documents use cases and test cases.

Tests

  • Bug fixes should have at least one unit or end-to-end test, enhancement and new features should have a sufficient number of tests.

Documentation

  • Apply the release notes (needs details) label if users need to know about this change.
  • New features should be documented.
  • There should be some code comments as to why things are implemented the way they are.
  • There should be a comment at the top of each new or modified test to explain what the test does.

New flags

  • Is this flag really necessary?
  • Flag names must be clear and intuitive, use dashes (-), and have a clear help text.

If a workflow is added or modified:

  • Each item in Jobs should be named in order to mark it as required.
  • If the workflow needs to be marked as required, the maintainer team must be notified.

Backward compatibility

  • Protobuf changes should be wire-compatible.
  • Changes to _vt tables and RPCs need to be backward compatible.
  • RPC changes should be compatible with vitess-operator
  • If a flag is removed, then it should also be removed from vitess-operator and arewefastyet, if used there.
  • vtctl command output order should be stable and awk-able.

@vitess-bot vitess-bot bot added NeedsBackportReason If backport labels have been applied to a PR, a justification is required NeedsDescriptionUpdate The description is not clear or comprehensive enough, and needs work NeedsIssue A linked issue is missing for this Pull Request NeedsWebsiteDocsUpdate What it says labels Dec 11, 2024
@github-actions github-actions bot added this to the v22.0.0 milestone Dec 11, 2024
@shlomi-noach
increase wait timeout
52c48df 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
wait for a substantial number of results
9812005 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
Many improvements:
e716c29 
- Improve `drain()` logic
- `drain()` returns workers error
- Wait for `drain()` to complete
- Parallel worker is a local variable
- Extract parallel counters outside loop
- Check workers error when replacing parallel worker
- Batch commits
- Parallel pool is a local variable
- isApplicable: optimize for when the worker is at head (skip mutex)
- isApplicable: dependency logic for batches commits: current event/s lastCommitted compared with existing workers/ sequenceNumber
- Worker: issue Rollback if terminated within transaction
- Worker: check context cancellation
- Worker: use `VEventType_UNKNOWN` as a serialization hint.

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
Decorate VEvent with MustSave
19e07b4 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
populate and use MustSave
b90f5e3 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
workers use vplayer's pos. Fix dbClient.Begin() in VEventType_ROW
587f929 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
generalize vplayer.updatePos, use by worker
11ed86d 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
addressing synchronization issues
478d8d6 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
VEventType: support PREVIOUS_GTIDS
c0035ee 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
produce and accept binlogdatapb.VEventType_PREVIOUS_GTIDS
4bfee23 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
more synchronization adjustements
0fde032 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
more concurrency fixes. Everything looks to be correct at this time.
be2acbc 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
more debug info
05438d9 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
remvoe debug info
7587329 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
stricter check for pos timestamp
7aa016e 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
support batched mode in parallel mode
99e8e59 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
a tiny bit of unit testing. Much more to go
905b991 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
check for worker error on availableWorker()
120ab81 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
check for posReached before applying any event
e054980 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
use worker's dbClient in vr.setState; use same context for worker poo…
b295b89 
…l drain()

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
setDBClientSettings for worker's dbclient
a75b425 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
ensure storing pos on EOF
b3e9fdc 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
only setDBClientSettings if filter != nil
79d07b3 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
apply SQL mode for workers dbclient
5f04217 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
close worker connections on drain (do not reset, do not recycle)
088bf7d 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
remove excessive applyFunc param
80142de 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
no need for drain() to close the conneciton, as it is already clodes …
9fe3a46 
…in runBlp()

Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
safer checks for EOF
e8679c6 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
do not commit is pos is reached
fd86928 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
restore original relay log buffering logic: wait for just one event
61c8f5f 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
safer checks for EOF
b204ca6 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
nicer spacing
7fcdf2a 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
validating vcopier injected datetime
b92fd55 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach
Merge branch 'main' into vplayer-parallel-applier
f67b653 
Signed-off-by: Shlomi Noach <2607934+shlomi-noach@users.noreply.github.com>
@shlomi-noach shlomi-noach mentioned this pull request Jan 26, 2025
Closed
5 tasks
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented Feb 26, 2025

This PR is being marked as stale because it has been open for 30 days with no activity. To rectify, you may do any of the following:

  • Push additional commits to the associated branch.
  • Remove the stale label.
  • Add a comment indicating why it is not stale.

If no action is taken within 7 days, this PR will be closed.

@github-actions github-actions bot added the Stale Marks PRs as stale after a period of inactivity, which are then closed after a grace period. label Feb 26, 2025
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented Mar 5, 2025

This PR was closed because it has been stale for 7 days with no activity.

@github-actions github-actions bot closed this Mar 5, 2025
@shlomi-noach shlomi-noach reopened this Mar 5, 2025
@shlomi-noach shlomi-noach removed the Stale Marks PRs as stale after a period of inactivity, which are then closed after a grace period. label Mar 5, 2025
@frouioui frouioui modified the milestones: v22.0.0, v23.0.0 Apr 1, 2025
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented May 5, 2025

This PR is being marked as stale because it has been open for 30 days with no activity. To rectify, you may do any of the following:

  • Push additional commits to the associated branch.
  • Remove the stale label.
  • Add a comment indicating why it is not stale.

If no action is taken within 7 days, this PR will be closed.

@github-actions github-actions bot added the Stale Marks PRs as stale after a period of inactivity, which are then closed after a grace period. label May 5, 2025
@github-actions
Copy link
Copy Markdown
Contributor

github-actions bot commented May 12, 2025

This PR was closed because it has been stale for 7 days with no activity.

@github-actions github-actions bot closed this May 12, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@harshit-gangal harshit-gangal Awaiting requested review from harshit-gangal harshit-gangal will be requested when the pull request is marked ready for review harshit-gangal is a code owner

@mattlord mattlord Awaiting requested review from mattlord mattlord will be requested when the pull request is marked ready for review mattlord is a code owner

@beingnoble03 beingnoble03 Awaiting requested review from beingnoble03 beingnoble03 will be requested when the pull request is marked ready for review beingnoble03 is a code owner

@frouioui frouioui Awaiting requested review from frouioui frouioui will be requested when the pull request is marked ready for review frouioui is a code owner

@rohit-nayak-ps rohit-nayak-ps Awaiting requested review from rohit-nayak-ps rohit-nayak-ps will be requested when the pull request is marked ready for review rohit-nayak-ps is a code owner

@systay systay Awaiting requested review from systay systay will be requested when the pull request is marked ready for review systay is a code owner

@timvaillancourt timvaillancourt Awaiting requested review from timvaillancourt timvaillancourt will be requested when the pull request is marked ready for review timvaillancourt is a code owner

Copilot code review Copilot Awaiting requested review from Copilot Copilot will automatically review once the pull request is marked ready for review

Assignees

No one assigned

Labels

Component: VReplication NeedsBackportReason If backport labels have been applied to a PR, a justification is required NeedsDescriptionUpdate The description is not clear or comprehensive enough, and needs work NeedsIssue A linked issue is missing for this Pull Request NeedsWebsiteDocsUpdate What it says Stale Marks PRs as stale after a period of inactivity, which are then closed after a grace period. Type: Performance

Projects

None yet

Milestone

v23.0.0

Development

Successfully merging this pull request may close these issues.

2 participants

@shlomi-noach @frouioui