[
https://issues.apache.org/jira/browse/BEAM-263?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15275690#comment-15275690
]
Davor Bonaci commented on BEAM-263:
-----------------------------------
I'd say this is beyond the scope of the Apache Beam project at present time.
To be able to run data-processing pipelines at scale, such a runner would have
to encompass the whole functionality of a data-processing backend. We don't aim
to do that -- instead, we rely on existing backends and services, like Google
Cloud Dataflow, Apache Flink or Apache Spark. At present, we don't plan to
increase scope to try to create our own, internal backend.
Make sense?
> Generate Processing Infrastructure Code and Code for Runners in High Level
> Languages
> ------------------------------------------------------------------------------------
>
> Key: BEAM-263
> URL: https://issues.apache.org/jira/browse/BEAM-263
> Project: Beam
> Issue Type: Improvement
> Components: runner-ideas
> Reporter: Suminda Dharmasena
>
> In addition you can auto generate processing infrastructure code in languages
> like C/C++, Java, Scala, C#, F#, Rust, Julia, Swift, D, Nim, etc.
> Auto generated code will be the runner itself and should not need any
> additional infrastructure to run.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
