[
https://issues.apache.org/jira/browse/YARN-6620?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16198224#comment-16198224
]
Sunil G commented on YARN-6620:
-------------------------------
Thanks [~leftnoteasy] for the patch. Generally I am fine with latest patch. I
could commit this tomorrow if there are no objection
Not very much related to patch, there are couple of other points to note:
# {{ResourceUtils.java}} is shared by client also. Hence once we make mandatory
resources like cpu mem and gpu, client will not get a chance to submit in GB's
for mem fo eg. Server could decode it at the gate and convert to MBs. But
client is loosing its chance for easy usage of resource units. I think this has
to be a separate jira. And I think there is a ticket for this, i ll find it.
# {{resource-types.xml}} is used to add new resources. GPU is mandatory like
cpu and mem, but I think resources like numa/fpga etc should follow the
YARN-3926 model, correct?. My point is that, we have new configs related to
GPUs here. I think for other resource specific configs, we could use
resource-types.xml or any new config xml to have more clarity and modularity.
> [YARN-6223] NM Java side code changes to support isolate GPU devices by using
> CGroups
> -------------------------------------------------------------------------------------
>
> Key: YARN-6620
> URL: https://issues.apache.org/jira/browse/YARN-6620
> Project: Hadoop YARN
> Issue Type: Sub-task
> Reporter: Wangda Tan
> Assignee: Wangda Tan
> Attachments: YARN-6620.001.patch, YARN-6620.002.patch,
> YARN-6620.003.patch, YARN-6620.004.patch, YARN-6620.005.patch,
> YARN-6620.006-WIP.patch, YARN-6620.007.patch, YARN-6620.008.patch,
> YARN-6620.009.patch, YARN-6620.010.patch, YARN-6620.011.patch,
> YARN-6620.012.patch, YARN-6620.013.patch, YARN-6620.014.patch,
> YARN-6620.015.patch, YARN-6620.016.patch, YARN-6620.017.patch
>
>
> This JIRA plan to add support of:
> 1) GPU configuration for NodeManagers
> 2) Isolation in CGroups. (Java side).
> 3) NM restart and recovery allocated GPU devices
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
