You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Apache Spark (JIRA)" <ji...@apache.org> on 2015/10/21 19:49:27 UTC
[jira] [Commented] (SPARK-8835) Provide pluggable Congestion
Strategies to deal with Streaming load
[ https://issues.apache.org/jira/browse/SPARK-8835?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14967542#comment-14967542 ]
Apache Spark commented on SPARK-8835:
-------------------------------------
User 'huitseeker' has created a pull request for this issue:
https://github.com/apache/spark/pull/9200
> Provide pluggable Congestion Strategies to deal with Streaming load
> -------------------------------------------------------------------
>
> Key: SPARK-8835
> URL: https://issues.apache.org/jira/browse/SPARK-8835
> Project: Spark
> Issue Type: Sub-task
> Components: Streaming
> Reporter: François Garillot
>
> Second part of [SPARK-7398|https://issues.apache.org/jira/browse/SPARK-7398] (which has an over-arching, high-level design doc).
> An (outdated) [PoC implementation|https://github.com/huitseeker/spark/tree/ReactiveStreamingBackPressureControl/] exists.
> Note that throttling in the current (Spark 1.5.0) way pushes down the (fast-overflowing) buffer of received elements down to the TCP stack, which then has to deal with the break. This can be witnessed here :
> https://github.com/skyluc/spark-streaming-testbed/tree/test-runs-006/test-runs-006
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org