PD performance slows down when scaling out 50 nodes simultaneously
Open
affects-8.5
report/customer
type/enhancement
- Dominant language
- Go
- Stars
- 1.2k
- Forks
- 783
- Avg merge
- 5d 21h
- Merged PRs (30d)
- 36
Description
## Bug Report
### What did you do?
Scale out 50 nodes.
### What did you expect to see?
No impact on PD performance.
### What did you see instead?
From the profile below, we can see that many CPUs were used to calculate the progress of preparing.
### What version of PD are you using (`pd-server -V`)?
v8.5.1
w/a is set max-store-preparing-time to 0 to stop the progress calculation.
Contributor guide
Assessment
This issue has not been assessed yet.