[Job](Fix)Improve Event Publishing with Timeout #45103
Merged
Conversation
This file contains 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
### Summary: This PR refines the publishEvent method to improve event publishing reliability by introducing a timeout mechanism and enhanced logging. The changes allow for a more responsive system when attempting to publish events to the disruptor, especially in cases where the ring buffer may not have sufficient capacity at the time. #### Timeout Implementation: A 1-second timeout (in nanoseconds) is set, after which the event publishing attempt will stop if the required capacity is not available. The timeout is tracked using System.nanoTime() for precise elapsed time measurement. #### Remaining Capacity Check: The method checks if the remainingCapacity() of the ring buffer is greater than 1 (this can be adjusted based on your capacity requirements). If enough capacity is available, the event is published; otherwise, it waits and retries.
|
Thank you for your contribution to Apache Doris. Please clearly describe your PR:
|
|
run buildall |
TPC-H: Total hot run time: 40121 ms |
TPC-DS: Total hot run time: 197667 ms |
ClickBench: Total hot run time: 32.64 s |
morningman
reviewed
Dec 6, 2024
| return true; | ||
|
|
||
| // Timeout reached without publishing the event | ||
| LOG.warn("Failed to publish event within the specified timeout (1 second)." |
There was a problem hiding this comment.
In log, it should be current size of the buffer, not remaining size of the buffer?
|
run buildall |
TPC-H: Total hot run time: 40031 ms |
TPC-DS: Total hot run time: 197618 ms |
ClickBench: Total hot run time: 33.23 s |
|
run buildall |
TPC-H: Total hot run time: 39932 ms |
TPC-DS: Total hot run time: 197458 ms |
ClickBench: Total hot run time: 32.36 s |
morrySnow
reviewed
Dec 9, 2024
fe/fe-core/src/main/java/org/apache/doris/job/disruptor/TaskDisruptor.java
Outdated
Show resolved
Hide resolved
…sruptor.java Co-authored-by: morrySnow <[email protected]>
|
run buildall |
TPC-H: Total hot run time: 39984 ms |
TPC-DS: Total hot run time: 196827 ms |
ClickBench: Total hot run time: 32.89 s |
|
run buildall |
TPC-H: Total hot run time: 39946 ms |
ClickBench: Total hot run time: 33.32 s |
morrySnow
previously approved these changes
Dec 11, 2024
github-actions
bot
added
the
approved
|
PR approved by at least one committer and no changes requested. |
|
PR approved by anyone and no changes requested. |
zddr
reviewed
Dec 11, 2024
| // Check if there is enough remaining capacity in the ring buffer | ||
| // Adjusting to check if the required capacity is available (instead of hardcoding 1) | ||
| if (disruptor.getRingBuffer().remainingCapacity() > 1) { | ||
| // Publish the event if there is enough capacity |
There was a problem hiding this comment.
Will there be concurrency? When making a judgment, it is greater than 1, and when publishing, there is no capacity left
There was a problem hiding this comment.
Currently, only the time wheel is in use, and since the time wheel itself is single-threaded, there are no issues.
| disruptor.getRingBuffer().publishEvent(eventTranslator, args); | ||
| if (LOG.isDebugEnabled()) { | ||
| LOG.debug("publishEvent success,the remaining buffer size is {}", | ||
| disruptor.getRingBuffer().remainingCapacity()); |
There was a problem hiding this comment.
What is the purpose of 'remainingCapacity'？
There was a problem hiding this comment.
Simply to add debug logs.
|
|
||
| public boolean addTask(AbstractTask task) { | ||
| try { | ||
| executor.execute(() -> runTask(task)); // 直接提交任务 |
github-actions
bot
removed
the
approved
|
run buildall |
zddr
approved these changes
Dec 11, 2024
TPC-H: Total hot run time: 40061 ms |
TPC-DS: Total hot run time: 191201 ms |
ClickBench: Total hot run time: 33.22 s |
zy-kkk
approved these changes
Dec 11, 2024
github-actions
bot
added
the
approved
|
PR approved by at least one committer and no changes requested. |
github-actions bot
pushed a commit
that referenced
this pull request
Dec 11, 2024
### Summary: This PR refines the publishEvent method to improve event publishing reliability by introducing a timeout mechanism and enhanced logging. The changes allow for a more responsive system when attempting to publish events to the disruptor, especially in cases where the ring buffer may not have sufficient capacity at the time. #### Timeout Implementation: A 1-second timeout (in nanoseconds) is set, after which the event publishing attempt will stop if the required capacity is not available. The timeout is tracked using System.nanoTime() for precise elapsed time measurement. #### Remaining Capacity Check: The method checks if the remainingCapacity() of the ring buffer is greater than 1 (this can be adjusted based on your capacity requirements). If enough capacity is available, the event is published; otherwise, it waits and retries.
github-actions bot
pushed a commit
that referenced
this pull request
Dec 11, 2024
### Summary: This PR refines the publishEvent method to improve event publishing reliability by introducing a timeout mechanism and enhanced logging. The changes allow for a more responsive system when attempting to publish events to the disruptor, especially in cases where the ring buffer may not have sufficient capacity at the time. #### Timeout Implementation: A 1-second timeout (in nanoseconds) is set, after which the event publishing attempt will stop if the required capacity is not available. The timeout is tracked using System.nanoTime() for precise elapsed time measurement. #### Remaining Capacity Check: The method checks if the remainingCapacity() of the ring buffer is greater than 1 (this can be adjusted based on your capacity requirements). If enough capacity is available, the event is published; otherwise, it waits and retries.
CalvinKirs
added a commit
that referenced
this pull request
Dec 12, 2024
…45300) Cherry-picked from #45103 Co-authored-by: Calvin Kirs <[email protected]>
CalvinKirs
added a commit
that referenced
this pull request
Dec 12, 2024
…45299) Cherry-picked from #45103 Co-authored-by: Calvin Kirs <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Summary:
This PR refines the publishEvent method to improve event publishing reliability by introducing a timeout mechanism and enhanced logging. The changes allow for a more responsive system when attempting to publish events to the disruptor, especially in cases where the ring buffer may not have sufficient capacity at the time.
Timeout Implementation:
A 1-second timeout (in nanoseconds) is set, after which the event publishing attempt will stop if the required capacity is not available. The timeout is tracked using System.nanoTime() for precise elapsed time measurement.
Remaining Capacity Check:
The method checks if the remainingCapacity() of the ring buffer is greater than 1 (this can be adjusted based on your capacity requirements). If enough capacity is available, the event is published; otherwise, it waits and retries.
Check List (For Author)
Test
Behavior changed:
Does this need documentation?
Check List (For Reviewer who merge this PR)