[
https://issues.apache.org/jira/browse/TEZ-1074?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Rajesh Balamohan updated TEZ-1074:
----------------------------------
Description:
- Ran a job which used 200 containers.
- DAGAppMaster was running at 70% most of the time during the job.
- Profiling revealed that lots of time was spent on TezEvent.readFields --> ...
--> TaskStatusUpdateEvent.readFields().
- Default "tez.task.am.heartbeat.interval-ms.max=100" ms. With 200 containers,
2000 events (these events have TezCounters) were processed by DAGAppMaster.
With large job, cpu usage can bloat up significantly.
One option to reduce CPU usage could be to send modified TezCounters in
TezStatusUpdateEvent.
was:
- Ran a job which used 200 containers.
- DAGAppMaster was running at 70% most of the time during the job.
- Profiling revealed that lots of time was spent on TezEvent.readFields --> ...
--> TaskStatusUpdateEvent.readFields().
- Default "tez.task.am.heartbeat.interval-ms.max=100" ms. With 200 containers,
2000 events (containing counters) were processed by DAGAppMaster.
With large job, cpu usage can bloat up significantly.
One option to reduce CPU usage could be to send modified TezCounters in
TezStatusUpdateEvent.
> DAGAppMaster takes lots of CPU when running a reasonably large job in the
> cluster
> ---------------------------------------------------------------------------------
>
> Key: TEZ-1074
> URL: https://issues.apache.org/jira/browse/TEZ-1074
> Project: Apache Tez
> Issue Type: Bug
> Reporter: Rajesh Balamohan
>
> - Ran a job which used 200 containers.
> - DAGAppMaster was running at 70% most of the time during the job.
> - Profiling revealed that lots of time was spent on TezEvent.readFields -->
> ... --> TaskStatusUpdateEvent.readFields().
> - Default "tez.task.am.heartbeat.interval-ms.max=100" ms. With 200
> containers, 2000 events (these events have TezCounters) were processed by
> DAGAppMaster.
> With large job, cpu usage can bloat up significantly.
> One option to reduce CPU usage could be to send modified TezCounters in
> TezStatusUpdateEvent.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
