[
https://issues.apache.org/jira/browse/CARBONDATA-830?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jacky Li resolved CARBONDATA-830.
---------------------------------
Resolution: Fixed
Fix Version/s: 1.1.0-incubating
> Incorrect schedule for NewCarbonDataLoadRDD
> -------------------------------------------
>
> Key: CARBONDATA-830
> URL: https://issues.apache.org/jira/browse/CARBONDATA-830
> Project: CarbonData
> Issue Type: Bug
> Components: spark-integration
> Affects Versions: 1.0.0-incubating
> Environment: Spark 2.1 + Carbon 1.0.0
> Reporter: Weizhong
> Assignee: Weizhong
> Priority: Minor
> Fix For: 1.1.0-incubating
>
> Time Spent: 1h
> Remaining Estimate: 0h
>
> Currently NewCarbonDataLoadRDD's getPreferredLocations will return all locs
> rather than 1, then on Spark may pick the same node for two tasks, so one
> node is getting over loaded with the task and one has no task to do, and
> impacting the performance despite of any failure.
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
