Github user chiwanpark commented on the pull request:
https://github.com/apache/flink/pull/696#issuecomment-131086738
@kno10 Thanks for the comment. In this case we don't need to parallelize
R-Tree because R-Tree is only used in reducer for matching records of the given
block pair.
But I agree that exact k-NN implementation doesn't fit large-scale data. We
can discuss in other [JIRA
issue](https://issues.apache.org/jira/browse/FLINK-1934).
I'm inclined to close this PR and FLINK-1745 because many people says exact
k-NN is not good and I think so. May I close them? @thvasilo @tillrohrmann
---
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.
---
