Github user rapen commented on the pull request:
https://github.com/apache/spark/pull/4449#issuecomment-108602224
@squito thanks! your work is very helpful. I'm testing now solution based
on your code. I've subclassed NewHadoopRDD with same changes you've made in
HaddopRDD and then created new method that creates this custom rdd(kind of
copy-paste from newHadoopApi from SparkContext) + defined NonSplittable
InputFormat(subclassed from avro formats). Thus I don't need to change
HadoopRDD and recompile spark with it...it's just sort of extension(might be
this what can be part of standalone library...I'm not pro in scala programming,
so not sure to show this code to someone :)
Anyway, all shuffles disappeared ! :+1:
---
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]
