You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "jincheng (JIRA)" <ji...@apache.org> on 2018/03/15 05:21:00 UTC

[jira] [Commented] (SPARK-20883) Improve StateStore APIs for efficiency

    [ https://issues.apache.org/jira/browse/SPARK-20883?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16399930#comment-16399930 ] 

jincheng commented on SPARK-20883:
----------------------------------

it is convenient to cherry pick to 2.2 without any changes.

> Improve StateStore APIs for efficiency
> --------------------------------------
>
>                 Key: SPARK-20883
>                 URL: https://issues.apache.org/jira/browse/SPARK-20883
>             Project: Spark
>          Issue Type: Improvement
>          Components: Structured Streaming
>    Affects Versions: 2.2.0
>            Reporter: Tathagata Das
>            Assignee: Tathagata Das
>            Priority: Major
>             Fix For: 2.3.0
>
>
> Current state store API has a bunch of problems that causes too many transient objects causing memory pressure.
> - StateStore.get() returns Options which forces creation of Some/None objects for every get
> - StateStore.iterator() returns tuples which forces creation of new tuple for each record returned
> - StateStore.updates() requires the implementation to keep track of updates, while this is used minimally (only by Append mode in streaming aggregations). This can be totally removed.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org