[
https://issues.apache.org/jira/browse/FLINK-38865?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18049946#comment-18049946
]
Vishal Kamlapure edited comment on FLINK-38865 at 1/5/26 9:33 PM:
------------------------------------------------------------------
I’m working on this issue and will submit a documentation-only PR.
was (Author: JIRAUSER311517):
I am working on the issue.
> Docs: Fix typos and clarify capacity scope in Async I/O documentation
> ---------------------------------------------------------------------
>
> Key: FLINK-38865
> URL: https://issues.apache.org/jira/browse/FLINK-38865
> Project: Flink
> Issue Type: Improvement
> Components: Documentation
> Reporter: Vishal Kamlapure
> Priority: Minor
>
> The Async I/O documentation contains a few minor issues that could be
> improved.
> 1) *Typo:* “This leads in most *cased* to …” should be “{*}cases{*}”.
> 2) *Grammar:* “well as handling …” should be “{*}as{*} well as handling …”.
> 3) *Clarity:* The description of the `{*}capacity{*}` parameter does not
> clearly state
> whether the limit applies per operator subtask or across all parallel
> subtasks.
> Currently, the paragraph explains the purpose of capacity, but it is unclear
> to readers
> whether capacity is enforced per subtask or globally.
> *Proposed improvement:*
> Explicitly state that capacity is applied per operator subtask, and that
> total concurrent
> async requests scale with operator parallelism.
> This is a documentation-only change.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
