crunch-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chao Shi (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (CRUNCH-307) Limit the number of concurrently running jobs
Date Wed, 27 Nov 2013 03:41:35 GMT

     [ https://issues.apache.org/jira/browse/CRUNCH-307?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Chao Shi updated CRUNCH-307:
----------------------------

    Attachment: crunch-307.patch

Here is my patch. I'm trying it on a real cluster.

> Limit the number of concurrently running jobs
> ---------------------------------------------
>
>                 Key: CRUNCH-307
>                 URL: https://issues.apache.org/jira/browse/CRUNCH-307
>             Project: Crunch
>          Issue Type: New Feature
>          Components: Core
>            Reporter: Chao Shi
>            Assignee: Chao Shi
>         Attachments: crunch-307.patch
>
>
> In current behavior, crunch keep submitting new jobs as long as their dependencies have
finished. This is not good on a scenario where a pipeline consists a large number (e.g. 10+)
of independent DAGs. A improvement is to define a limited window of concurrent running jobs.
Whenever such limit is reached, we stop submitting new jobs and wait until any running job
finishes.



--
This message was sent by Atlassian JIRA
(v6.1#6144)

Mime
View raw message