[
https://issues.apache.org/jira/browse/TEZ-2872?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14950675#comment-14950675
]
Bikas Saha commented on TEZ-2872:
---------------------------------
The downside of TEZ-307 is that a subsequent DAG will not be able to reuse
containers since the DAGs local resources would not match those of the previous
DAG. There is built-in relocalization in Tez for such cases, but thats not
scalable since every container tries to relocalize resources (instead of once
per job). But that might not be a problem in your case since you are mostly
running 1 large DAG per job (instead of a pig shell with multiple submissions)
right?
> Tez AM can be overwhelmed by TezTaskUmbilicalProtocol.getTask responses
> -----------------------------------------------------------------------
>
> Key: TEZ-2872
> URL: https://issues.apache.org/jira/browse/TEZ-2872
> Project: Apache Tez
> Issue Type: Bug
> Reporter: Jason Lowe
>
> When a large job runs on a large cluster with a large user payload then the
> AM can end up hitting OOM conditions. For example, Pig-on-Tez can require a
> significant user payload (approaching 1MB) for vertices, inputs, and outputs
> in the DAG. This can cause the ContainerTask response to be rather large per
> task, which can lead to a situation where the AM is generating output faster
> than the network interface can process it. If there are enough containers
> asking for tasks then this leads to an OOM condition.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
