Sangjin Lee commented on YARN-2928:

bq. We will have to carve out some capacity for the per-node companions. I see 
some sort of static allocation like 1GB similar to NodeManager.

The required memory for the per-node aggregator might be larger than 
anticipated. One reference point may be the memory footprint of a MR AM. The 
bulk of the job-related pinned-down memory would be needed on the aggregator. 
And that can be easily in the several hundreds of MB. Also, for buffering 
multiple of such data for writes would require more room. On top of that, one 
would need to multiply by the number of apps it needs to support (x2 or x3 at 

All in all, my gut feeling is that 1 GB might be rather tight. I think we'll 
know more as we start testing it with realistic size apps and backing storage.

> Application Timeline Server (ATS) next gen: phase 1
> ---------------------------------------------------
>                 Key: YARN-2928
>                 URL: https://issues.apache.org/jira/browse/YARN-2928
>             Project: Hadoop YARN
>          Issue Type: New Feature
>          Components: timelineserver
>            Reporter: Sangjin Lee
>            Assignee: Vinod Kumar Vavilapalli
>            Priority: Critical
>         Attachments: ATSv2.rev1.pdf, ATSv2.rev2.pdf
> We have the application timeline server implemented in yarn per YARN-1530 and 
> YARN-321. Although it is a great feature, we have recognized several critical 
> issues and features that need to be addressed.
> This JIRA proposes the design and implementation changes to address those. 
> This is phase 1 of this effort.

This message was sent by Atlassian JIRA

Reply via email to