[
https://issues.apache.org/jira/browse/YARN-11105?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
zhangjunj updated YARN-11105:
-----------------------------
Description:
Now YARN Federation does not support the running of jobs across DCs ( data
centers), because of dedicated line bandwidth and job processing latency, as
[YARN
Federation|https://hadoop.apache.org/docs/stable/hadoop-yarn/hadoop-yarn-site/Federation.html#:~:text=We%20assume%20reasonably%20good%20connectivity%20across%20sub%2Dclusters%20(e.g.%2C%20we%20are%20not%20looking%20to%20federate%20across%20DC%20yet%2C%20though%20future%20investigations%20of%20this%20are%20not%20excluded).]
said.
So I want to know, when will this feature be supported, and how to implement
jobs that only run under each DCs.
I found that this is not just my own needs, including other manufacturers I
know, and
[LinkIn|https://engineering.linkedin.com/blog/2021/scaling-linkedin-s-hadoop-yarn-cluster-beyond-10-000-nodes#:~:text=L7%20load%20balancer.-,Robin,-We%20envision%20our].
was:
Now YARN Federation does not support the running of jobs across DCs, because of
dedicated line bandwidth and job processing latency, as [YARN
Federation|https://hadoop.apache.org/docs/stable/hadoop-yarn/hadoop-yarn-site/Federation.html#:~:text=We%20assume%20reasonably%20good%20connectivity%20across%20sub%2Dclusters%20(e.g.%2C%20we%20are%20not%20looking%20to%20federate%20across%20DC%20yet%2C%20though%20future%20investigations%20of%20this%20are%20not%20excluded).]
said.
So I want to know, when will this feature be supported, and how to implement
jobs that only run under each DCs.
I found that this is not just my own needs, including other manufacturers I
know, and
[LinkIn|https://engineering.linkedin.com/blog/2021/scaling-linkedin-s-hadoop-yarn-cluster-beyond-10-000-nodes#:~:text=L7%20load%20balancer.-,Robin,-We%20envision%20our].
> Prohibit jobs from running across DC by YARN Federation
> -------------------------------------------------------
>
> Key: YARN-11105
> URL: https://issues.apache.org/jira/browse/YARN-11105
> Project: Hadoop YARN
> Issue Type: Wish
> Reporter: zhangjunj
> Priority: Major
>
> Now YARN Federation does not support the running of jobs across DCs ( data
> centers), because of dedicated line bandwidth and job processing latency, as
> [YARN
> Federation|https://hadoop.apache.org/docs/stable/hadoop-yarn/hadoop-yarn-site/Federation.html#:~:text=We%20assume%20reasonably%20good%20connectivity%20across%20sub%2Dclusters%20(e.g.%2C%20we%20are%20not%20looking%20to%20federate%20across%20DC%20yet%2C%20though%20future%20investigations%20of%20this%20are%20not%20excluded).]
> said.
> So I want to know, when will this feature be supported, and how to implement
> jobs that only run under each DCs.
> I found that this is not just my own needs, including other manufacturers I
> know, and
> [LinkIn|https://engineering.linkedin.com/blog/2021/scaling-linkedin-s-hadoop-yarn-cluster-beyond-10-000-nodes#:~:text=L7%20load%20balancer.-,Robin,-We%20envision%20our].
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: yarn-issues-h...@hadoop.apache.org
