[
https://issues.apache.org/jira/browse/FLINK-8144?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16265088#comment-16265088
]
Fabian Hueske commented on FLINK-8144:
--------------------------------------
Did you observe this behavior in practice, i.e., while debugging, or is that an
assumption based on reading the code?
I'm asking, because a timer that is registered on {{current_watermark + 1}}
fires just once when the next watermark is received (the logical clock is only
advanced by watermarks).
Also by registering multiple timers on the same timestamp, the timer gets
overridden, so there will be only one timer that fires.
I think the current implementation should behave just as expected, i.e., just
go once over the MapState when a watermark is received.
> Optimize the timer logic in RowTimeUnboundedOver
> ------------------------------------------------
>
> Key: FLINK-8144
> URL: https://issues.apache.org/jira/browse/FLINK-8144
> Project: Flink
> Issue Type: Bug
> Components: Table API & SQL
> Reporter: Dian Fu
> Assignee: Dian Fu
> Fix For: 1.5.0
>
>
> Currently the logic of {{RowTimeUnboundedOver}} is as follows:
> 1) When element comes, buffer it in MapState and and register a timer at
> {{current watermark + 1}}
> 2) When event timer triggered, scan the MapState and find the elements below
> the current watermark and process it. If there are remaining elements to
> process, register a new timer at {{current watermark + 1}}.
> Let's assume that watermark comes about 5 seconds later than the event on
> average, then we will scan about 5000 times the MapState before actually
> processing the events.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
