-
Notifications
You must be signed in to change notification settings - Fork 3.5k
[fix](replica num) Fix the decrease in the number of replicas and une… #48704
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.
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
Conversation
…ven distribution of replicas among bes
Thank you for your contribution to Apache Doris. Please clearly describe your PR:
|
run buildall |
TPC-H: Total hot run time: 32472 ms
|
TPC-DS: Total hot run time: 191447 ms
|
ClickBench: Total hot run time: 30.3 s
|
fe/fe-core/src/test/java/org/apache/doris/clone/DecreaseReplicationNumTest.java
Show resolved
Hide resolved
fe/fe-core/src/test/java/org/apache/doris/clone/DecreaseReplicationNumTest.java
Show resolved
Hide resolved
fe/fe-core/src/test/java/org/apache/doris/clone/DecreaseReplicationNumTest.java
Show resolved
Hide resolved
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
need update
run buildall |
run buildall |
TPC-H: Total hot run time: 32381 ms
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR approved by anyone and no changes requested. |
TPC-DS: Total hot run time: 185112 ms
|
ClickBench: Total hot run time: 30.75 s
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
PR approved by at least one committer and no changes requested. |
apache#48704) …ven distribution of replicas among bes
…ven distribution of replicas among bes
What problem does this PR solve?
Issue Number: close #xxx
Related PR: #xxx
Problem Summary:
When reducing the number of table replicas, the decision to drop replicas is currently based on the load situation of the BE nodes. However, this approach can result in the node with high BE load dropping many replicas at the same time, leading to severe CPU imbalance in the BE cluster.
The fix is to count the distributed tablet numbers on each BE when altering the number of replicas. Based on this mapping, we will determine which BEs to drop replicas from.
Release note
None
Check List (For Author)
Test
Behavior changed:
Does this need documentation?
Check List (For Reviewer who merge this PR)