[jira] [Commented] (SPARK-11384) Wrong duration times for (cached) iterative tasks

["Theodore Vasiloudis (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-11384?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14978901#comment-14978901 ] 

Theodore Vasiloudis commented on SPARK-11384:
---------------------------------------------

If you look at the screenshot and focus on the duration for the task that reads Cluster:164.

Even though the total job runtime at the time I took the screenshot was ~53min that particular task has its duration reported as the total runtime until that time, instead of the actual time it took to run.

That task involves a cached RDD, but I'm not sure if that is relevant.

Is this the intended behavior? Shouldn't each task report the time it took to run individually?

> Wrong duration times for (cached) iterative tasks
> -------------------------------------------------
>
>                 Key: SPARK-11384
>                 URL: https://issues.apache.org/jira/browse/SPARK-11384
>             Project: Spark
>          Issue Type: Bug
>          Components: Web UI
>    Affects Versions: 1.5.0
>            Reporter: Theodore Vasiloudis
>            Priority: Minor
>         Attachments: Web UI tasks.png
>
>
> In the Web UI when I'm running a job that includes an iteration, some stages that include an RDD that I cache appear multiple times, and it seems their duration is just incremented instead of reset at the beginning of each iteration.
> I will try to attach a screenshot to better explain this.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org