[jira] [Closed] (FLINK-6359) Utilize Hierarchical Timing Wheels for performant timer

["Stephan Ewen (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/FLINK-6359?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Stephan Ewen closed FLINK-6359.
-------------------------------
    Resolution: Won't Do

Closing this for now, because this issue is inactive and does not have any applicable design proposal, yet.

The Flink timers need to be persistent, out-of-core, and rescalable (by key) which are requirements not given by the time wheel structures.

> Utilize Hierarchical Timing Wheels for performant timer
> -------------------------------------------------------
>
>                 Key: FLINK-6359
>                 URL: https://issues.apache.org/jira/browse/FLINK-6359
>             Project: Flink
>          Issue Type: Improvement
>          Components: API / DataStream, Runtime / Task
>            Reporter: Ted Yu
>            Priority: Major
>              Labels: perfomance
>
> In this thread on mailing list:
> http://search-hadoop.com/m/Flink/VkLeQPmRa31hd5cw
> Gyula Fóra mentioned that timer deletion becomes performance bottleneck due to the usage of priority queue.
> Benjamin has an implementation for Hierarchical Timing Wheels (Apache License) :
> https://github.com/ben-manes/caffeine/blob/master/caffeine/src/main/java/com/github/benmanes/caffeine/cache/TimerWheel.java
> {code}
>  * A hierarchical timer wheel to add, remove, and fire expiration events in amortized O(1) time. The
>  * expiration events are deferred until the timer is advanced, which is performed as part of the
>  * cache's maintenance cycle.
> {code}
> We should consider porting the above over to facilitate performant timer.



--
This message was sent by Atlassian Jira
(v8.3.4#803005)