flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Fabian Hueske (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (FLINK-170) Introduce manual memory configuration
Date Mon, 08 Sep 2014 23:00:35 GMT

     [ https://issues.apache.org/jira/browse/FLINK-170?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Fabian Hueske updated FLINK-170:
--------------------------------
    Priority: Minor  (was: Major)

> Introduce manual memory configuration
> -------------------------------------
>
>                 Key: FLINK-170
>                 URL: https://issues.apache.org/jira/browse/FLINK-170
>             Project: Flink
>          Issue Type: Improvement
>            Reporter: GitHub Import
>            Priority: Minor
>              Labels: github-import
>             Fix For: pre-apache
>
>
> It is currently not possible to influence the optimizer/compiler in its decision to distribute
the memory among the different tasks.
> A way to weight the different operators, or to prioritize them somehow would greatly
help to improve the memory utilization (in terms of performance gain).
> Probably we could also add an overbooking of memory.
> ---------------- Imported from GitHub ----------------
> Url: https://github.com/stratosphere/stratosphere/issues/170
> Created by: [rmetzger|https://github.com/rmetzger]
> Labels: core, enhancement, optimizer, 
> Assignee: [StephanEwen|https://github.com/StephanEwen]
> Created at: Wed Oct 16 17:44:56 CEST 2013
> State: open



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message