[
https://issues.apache.org/jira/browse/TAJO-603?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Min Zhou updated TAJO-603:
--------------------------
Attachment: schedule.png
Here is a state diagram showing how DefaultTaskScheduler handle the events and
to hold the scheduled task requests. You can observe that there is a event
circulation. Event fired from DefaultTaskScheduler, send to QueryUnit and then
QueryUnitAttempt, finally the DefaultTaskScheduler handles the the same event
(just different type) it fires.
See the code in DefaultTaskScheduler.java
{noformat}
public void handle(TaskSchedulerEvent event) {
if (event.getType() == EventType.T_SCHEDULE) {
...
} else if (event instanceof FetchScheduleEvent) {
...
} else if (event instanceof QueryUnitAttemptScheduleEvent) {
...
}
}
}
{noformat}
> Move container allocation from SubQuery down to QueryUnitAttempt
> ----------------------------------------------------------------
>
> Key: TAJO-603
> URL: https://issues.apache.org/jira/browse/TAJO-603
> Project: Tajo
> Issue Type: Sub-task
> Reporter: Min Zhou
> Attachments: schedule.png
>
>
> Tajo currently allocates all of the containers in SubQuery. That make things
> complicated. Both SubQuery and DefaultTaskScheduler should hold a copy of
> allocated containers and running tasks. And the event flow is difficult to
> understand.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)
