Skip to content
This repository has been archived by the owner on Jun 23, 2022. It is now read-only.

feat(bulk_load): meta server adds bulk load ingestion concurrent count restriction #829

Merged
merged 10 commits into from
May 7, 2021

Conversation

hycdong
Copy link
Contributor

@hycdong hycdong commented Apr 30, 2021

To shorten all partitions ingestion total time, the pervious implementation didn't do any restrction when meta server sent ingestion request to primaries. However, ingestion will cost replica server lots of cpu, io resource, even affect cluster read/write performance. This pull request adds ingestion concurrent count in meta server, if the count is 4, the fifth ingestion request won't be sent to primary partition unless meta received one ingestion response. Besides, this count can be updated dynamically through http interface.

[meta_server]
+bulk_load_ingestion_concurrent_count=4

@hycdong hycdong added component/bulk-load type/config-change PR that made modification on configs, which should be noted in release note. labels Apr 30, 2021
@hycdong hycdong changed the title feat(bulk_load): meta server add bulk load ingestion concurrent count restriction feat(bulk_load): meta server adds bulk load ingestion concurrent count restriction Apr 30, 2021
@hycdong hycdong marked this pull request as draft April 30, 2021 07:01
@hycdong hycdong marked this pull request as ready for review May 6, 2021 07:36
@hycdong hycdong merged commit 6693cf7 into XiaoMi:master May 7, 2021
@hycdong hycdong deleted the bulk_load_ingestion branch May 7, 2021 03:50
hycdong added a commit that referenced this pull request May 7, 2021
hycdong added a commit that referenced this pull request May 25, 2021
@hycdong hycdong added the 2.2.0 label Aug 30, 2021
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
2.2.0 component/bulk-load type/config-change PR that made modification on configs, which should be noted in release note.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants