Github user nchammas commented on the pull request:
https://github.com/apache/spark/pull/3564#issuecomment-67786877
Yeah, I looked briefly at ways of parallelizing the Python tests, because
those take around 10-12 minutes in total, which will become a significant
fraction of the total test time once the Scala tests can run in parallel.
I looked at using GNU parallel, but there seem to be some annoying things
related to copying environment configs and whatnot to get that working. A more
promising approach I think would be to look into using
[`nose`](http://nose.readthedocs.org/en/latest/doc_tests/test_multiprocess/multiprocess.html),
which extends `unittest` which we are using, to parallelize the Python tests.
In that case, yeah, I think it might make sense to move PyPy builds out of
the pull request builder until we containerize the build/test cycle on Jenkins.
With containerization, we should be able to look into having everything build
in parallel with one container per "first-class" configuration (e.g. Hadoop
1/2, Scala 2.10/2.11, Java 7/8, Python 2/3, CPython/PyPy, etc.) without
worrying about contention over resources like ports or files.
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
