[Feature] Maintain the limited sorted runs in write-only mode
- Dominant language
- Java
- Stars
- 3.4k
- Forks
- 1.4k
- Avg merge
- 1d 11h
- Merged PRs (30d)
- 396
Description
### Search before asking
- [X] I searched in the [issues](https://github.com/apache/incubator-paimon/issues) and found nothing similar.
### Motivation
Currently, when we using `write-only` mode with a dedicated compaction job, then the frontend write job's speed are not limited by the compaction job. It may lead to mismatch speed of the whole pipeline. This will lead to much merge read process during query. So, I think a Write-Stall for the multi write job if the sorted runs exceed the `num-sorted-run.stop-trigger` may do good to the whole pipeline.
### Solution
_No response_
### Anything else?
_No response_
### Are you willing to submit a PR?
- [X] I'm willing to submit a PR!
Contributor guide
No contributing guide indexed for this repository
Research direction
Start by tracing the write-only mode and the handling of sorted runs, then locate where num-sorted-run.stop-trigger is applied. Determine how a write stall should coordinate with a dedicated compaction job and multiple writers; done means writes are limited when sorted runs exceed the configured trigger, preventing unbounded merge-read growth.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- java
- Domain
- data-engineering, databases
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100