Skip to content

[FLINK-14747] Fix performance regression with latest changes to Mailbox #10177

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
pnowojski merged 5 commits into from
Nov 21, 2019
Merged

[FLINK-14747] Fix performance regression with latest changes to Mailbox #10177

pnowojski merged 5 commits into from
Nov 21, 2019

Conversation

@AHeise
Copy link
Contributor

@AHeise AHeise commented Nov 13, 2019

What is the purpose of the change

  • Fixes performance regression introduced with FLINK-14304

Brief change log

  • Reintroduce volatile variable to avoid lock acquisition on empty queue.

Verifying this change

(Please pick either of the following options)

  • Correctness covered by existing tests.
  • Performance visible in benchmarks.

Does this pull request potentially affect one of the following parts:

  • Dependencies (does it add or upgrade a dependency): (no)
  • The public API, i.e., is any changed class annotated with @Public(Evolving): (no)
  • The serializers: (no)
  • The runtime per-record code paths (performance sensitive): (yes)
  • Anything that affects deployment or recovery: JobManager (and its components), Checkpointing, Yarn/Mesos, ZooKeeper: (no)
  • The S3 file system connector: (no)

Documentation

  • Does this pull request introduce a new feature? (no)
  • If yes, how is the feature documented? (not applicable)

@flinkbot
Copy link
Collaborator

flinkbot commented Nov 13, 2019 •
edited
Loading

Thanks a lot for your contribution to the Apache Flink project. I'm the @flinkbot. I help the community
to review your pull request. We will use this comment to track the progress of the review.

Automated Checks

Last check on commit 70195de (Wed Apr 15 11:37:09 UTC 2020)

Warnings:

  • No documentation files were touched! Remember to keep the Flink docs up to date!

Mention the bot in a comment to re-run the automated checks.

Review Progress

  • ❓ 1. The [description] looks good.
  • ❓ 2. There is [consensus] that the contribution should go into to Flink.
  • ❓ 3. Needs [attention] from.
  • ❓ 4. The change fits into the overall [architecture].
  • ❓ 5. Overall code [quality] is good.

Please see the Pull Request Review Guide for a full explanation of the review process.

Details
The Bot is tracking the review progress through labels. Labels are applied according to the order of the review items. For consensus, approval by a Flink committer of PMC member is required Bot commands
The @flinkbot bot supports the following commands:

  • @flinkbot approve description to approve one or more aspects (aspects: description, consensus, architecture and quality)
  • @flinkbot approve all to approve all aspects
  • @flinkbot approve-until architecture to approve everything until architecture
  • @flinkbot attention @username1 [@username2 ..] to require somebody's attention
  • @flinkbot disapprove architecture to remove an approval you gave earlier

@flinkbot
Copy link
Collaborator

flinkbot commented Nov 13, 2019 •
edited
Loading

CI report:

Bot commands The @flinkbot bot supports the following commands:
  • @flinkbot run travis re-run the last Travis build

pnowojski
pnowojski requested changes Nov 14, 2019
View reviewed changes
Copy link
Contributor

@pnowojski pnowojski left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for quickly identifying the issue and fixing it. I've left couple of comments

...ing-java/src/main/java/org/apache/flink/streaming/runtime/tasks/mailbox/TaskMailboxImpl.java

@Nullable
private Mail takeOrNull(Deque<Mail> queue, int priority) {
checkTakeStateConditions();
Copy link
Contributor

@pnowojski pnowojski Nov 14, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

why was this removed? Is it related to the performance fix?

Copy link
Contributor

@pnowojski pnowojski Nov 19, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bump?

Copy link
Contributor Author

@AHeise AHeise Nov 19, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It's related to the correct behavior of take and tryTake. Internally, they know invoke that function takeOrNull twice, which would result in two state checks. By moving that to the callee, it's now a mini-tad faster ;).

Copy link
Contributor

@pnowojski pnowojski Nov 20, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, I see now :)

@AHeise AHeise requested a review from pnowojski November 19, 2019 13:06
pnowojski
pnowojski reviewed Nov 19, 2019
View reviewed changes
Copy link
Contributor

@pnowojski pnowojski left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks almost good to me. I think one comment might have been not addressed.

...ing-java/src/main/java/org/apache/flink/streaming/runtime/tasks/mailbox/TaskMailboxImpl.java

@Nullable
private Mail takeOrNull(Deque<Mail> queue, int priority) {
checkTakeStateConditions();
Copy link
Contributor

@pnowojski pnowojski Nov 19, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

bump?

pnowojski
pnowojski approved these changes Nov 20, 2019
View reviewed changes
...ing-java/src/main/java/org/apache/flink/streaming/runtime/tasks/mailbox/TaskMailboxImpl.java

@Nullable
private Mail takeOrNull(Deque<Mail> queue, int priority) {
checkTakeStateConditions();
Copy link
Contributor

@pnowojski pnowojski Nov 20, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, I see now :)

pnowojski
pnowojski requested changes Nov 20, 2019
View reviewed changes
Copy link
Contributor

@pnowojski pnowojski left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@AHeise there seems to be some test failures, including deadlocks.

Arvid Heise added 5 commits November 20, 2019 11:50
[FLINK-14747][runtime] Reintroducing volatile flag in TaskMailboxImpl to
9fcc20b 
    avoid locks in case of empty mail queue.
[hotfix][runtime] MockStreamTask gracefully stops mailbox processor.
110f07a 
MockStreamTask cannot call close on mailbox processor as we cannot
assume that #cleanup is called from mailbox thread. Using a graceful
termination allows the stream task and any using test harness to be
closed from another thread for failure simulation.
[hotfix][runtime] Smaller cleanup of Mailbox and ensuring state changing
2df207b 
operation may only occur through the mailbox thread, such that state
checks do not any kind of synchronizations.
[hotfix][runtime] Avoid iteration creation in
378b366 
TaskMailboxImpl#tryTakeFromBatch.
[hotfix][runtime] Fixing TaskMailboxImpl #take and #tryTake not honoring
70195de 
requested priority if a batch is open.
pnowojski
pnowojski approved these changes Nov 21, 2019
View reviewed changes
Copy link
Contributor

@pnowojski pnowojski left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for fixing the tests! Merging.

@pnowojski pnowojski merged commit cc309c5 into apache:master Nov 21, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@pnowojski pnowojski pnowojski approved these changes

Assignees

No one assigned

Labels

component=Runtime/Task review=description?

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@AHeise @flinkbot @pnowojski @rmetzger