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

Posted to user@ignite.apache.org by Nikolai Tikhonov <nt...@gridgain.com> on 2017/05/01 10:33:17 UTC

Re: embedded mode vs standalone mode

Hi,

> In the embedded mode, can the "state" be shared across multiple Spark
> applications, if the Ignite server within the Spark application not die? 

In embedded mode applications can share the state but only during lifespan.

>If the "state" can only be shared in standalone mode, what is the value of
embedded mode? 

In standalone mode you can use IgniteRDD and read data from Ignite. 




--
View this message in context: http://apache-ignite-users.70518.x6.nabble.com/embedded-mode-vs-standalone-mode-tp12317p12319.html
Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: embedded mode vs standalone mode

Posted by Nikolai Tikhonov <nt...@apache.org>.

In this case you need to use standalone mode. In first job you need to
store data to IgniteRDD by IgniteRDD#saveValues. In second job you shoud
create RDD from the same cache and result of first job will be avaliable
threre.

On Mon, May 1, 2017 at 1:33 PM, Nikolai Tikhonov <nt...@gridgain.com>
wrote:

> Hi,
>
> > In the embedded mode, can the "state" be shared across multiple Spark
> > applications, if the Ignite server within the Spark application not die?
>
> In embedded mode applications can share the state but only during lifespan.
>
> >If the "state" can only be shared in standalone mode, what is the value of
> embedded mode?
>
> In standalone mode you can use IgniteRDD and read data from Ignite.
>
>
>
>
> --
> View this message in context: http://apache-ignite-users.
> 70518.x6.nabble.com/embedded-mode-vs-standalone-mode-tp12317p12319.html
> Sent from the Apache Ignite Users mailing list archive at Nabble.com.
>