You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Apache Spark (Jira)" <ji...@apache.org> on 2021/01/15 06:13:00 UTC
[jira] [Commented] (SPARK-34123) Faster way to display/render
entries in HistoryPage (Spark history server summary page)
[ https://issues.apache.org/jira/browse/SPARK-34123?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17265730#comment-17265730 ]
Apache Spark commented on SPARK-34123:
--------------------------------------
User 'mohan3d' has created a pull request for this issue:
https://github.com/apache/spark/pull/31191
> Faster way to display/render entries in HistoryPage (Spark history server summary page)
> ---------------------------------------------------------------------------------------
>
> Key: SPARK-34123
> URL: https://issues.apache.org/jira/browse/SPARK-34123
> Project: Spark
> Issue Type: Improvement
> Components: Web UI
> Affects Versions: 3.2.0
> Reporter: Mohanad Elsafty
> Priority: Major
> Attachments: Screenshot 2021-01-15 at 1.21.40 PM.png
>
>
> Since a long time ago my team/company suffered from history server being very slow to display/search entries specially when entries grow over 50k entry, regardless there is a pagination there in that page already but still very slow to display the entries.
>
> Current situation *Mustache Js* is used to render the entries and *datatables* is used to manipulate it (sort by column and search).
>
> By getting rid of *Mustache* (stop rendering the entries using *Mustache*) and using *datatables* to display it proved to be faster.
>
> Displaying > 100k entries (my case):
> Existing takes at least 30 to 40 seconds to display the entries, searching takes at least 20 seconds and the page stop responding until it finishes.
> Improved takes ~3 seconds to display the entries searching is very fast and the page stays responsive.
> *(These numbers will be different for others since JS is executed on your browser)*
>
> I am not sure why *Mustache* is used to display the data since data tables can do the job,
> [~ajbozarth] [~sowen] please elaborate more about this what is the reason to use *Mustache*? what are the drawbacks if it's not used anymore to display the entries (only this part)?
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org