[
https://issues.apache.org/jira/browse/FLINK-22830?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17358565#comment-17358565
]
Arvid Heise commented on FLINK-22830:
-------------------------------------
I'm responding on stackoverflow. Such question are best asked on user [mailing
list|https://flink.apache.org/gettinghelp.html]. A ticket is not meant for
asking support questions.
> DataSet API to allow to collect and do the bulk upload in DB
> ------------------------------------------------------------
>
> Key: FLINK-22830
> URL: https://issues.apache.org/jira/browse/FLINK-22830
> Project: Flink
> Issue Type: Improvement
> Components: API / DataSet
> Reporter: Ravi
> Priority: Major
>
> Experts,
>
> I am trying to perform some ETL operation on large data set as batch
> processing. My requirement is to extract the data , transform it and then
> save to mongoDB. I am using Apache FLINK but the performance is very slow as
> I am doing the mongoDB update on each row.
> Is there any way where we can sink as bulk record so that performance can
> increase. Like after all the transformation we do the bulk update on mongoDB.
> We can aggregate them all and finally sink it in DB just like stream
> [.aggregate() .sink(\{bulk update})] . Please refer
> [https://stackoverflow.com/questions/67717964/java-apache-flink-batch-processing-performance-for-bulk-mongodb-update]
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
