Github user pwendell commented on the pull request:
https://github.com/apache/spark/pull/26#issuecomment-36278255
Hey @ngbinh this is a cool feature, but it might be best as a standalone
project that we can refer to on the Spark website or in a separate contrib repo
(similar to https://github.com/nathanmarz/storm-contrib - an idea we've been
kicking around). The only thing really similar to this is the EC2 scripts,
which are actually hosted largely outside of Spark's own repo, and we are
thinking of further de-coupling them from the spark codebase.
Given that the number of ways people deploy spark is proliferating, I think
the project is best off pointing people towards other maintained repositories
rather than embedding and supporting a bunch of different ways of doing this.
For instance, a while before there was a proposed 1000-line patch to add
openstack support:
https://github.com/apache/incubator-spark/pull/91
And offline people have talked to me about contributing patches for docker
(a user facing docker thing, not just one we use for tests), google compute
engine, and other deployment modes.
Merging all of these into Spark would make the project unwieldy. However,
pointing people towards them would be great. So if you are interested in
maintaining this as its own repo, we could probably codify that and point users
towards it.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
