You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@spark.apache.org by Ashish Soni <as...@gmail.com> on 2015/07/01 20:14:35 UTC

BroadCast Multiple DataFrame ( JDBC Tables )

Hi ,

I need to load 10 tables in memory and have them available to all the
workers , Please let me me know what is the best way to do broadcast them

sc.broadcast(df)  allow only one

Thanks,

Re: BroadCast Multiple DataFrame ( JDBC Tables )

Posted by Raghavendra Pandey <ra...@gmail.com>.

I am not sure if you can broadcast data frame without collecting it on
driver...
On Jul 1, 2015 11:45 PM, "Ashish Soni" <as...@gmail.com> wrote:

> Hi ,
>
> I need to load 10 tables in memory and have them available to all the
> workers , Please let me me know what is the best way to do broadcast them
>
> sc.broadcast(df)  allow only one
>
> Thanks,
>
>
>