[
https://issues.apache.org/jira/browse/BEAM-260?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15964694#comment-15964694
]
Thomas Groh commented on BEAM-260:
----------------------------------
This is supported in the Java SDK. I've filed
https://issues.apache.org/jira/browse/BEAM-1931 and
https://issues.apache.org/jira/browse/BEAM-1932 to track the use of this in the
Java SDK, and I'm going to close this as preempted by those two issues.
> WindowMappingFn: Know the getSideInputWindow upper bound to release side
> input resources
> ----------------------------------------------------------------------------------------
>
> Key: BEAM-260
> URL: https://issues.apache.org/jira/browse/BEAM-260
> Project: Beam
> Issue Type: Bug
> Components: beam-model-fn-api, beam-model-runner-api
> Reporter: Mark Shields
> Assignee: Thomas Groh
>
> We currently have no static knowledge about the getSideInputWindow function,
> and runners are thus forced to hold on to all side input state / elements in
> case a future element reaches back into an earlier side input element.
> Maybe we need an upper bound on lag from current to result of
> getSideInputWindow so we can have a progressing gc horizon as we do for GKB
> window state.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
