Github user vasia commented on the pull request:
https://github.com/apache/flink/pull/1671#issuecomment-215506281
I see. Thanks for the explanation!
---
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
Github user greghogan commented on the pull request:
https://github.com/apache/flink/pull/1671#issuecomment-215503980
The two implementations have small differences but the algorithm is the
same. I'll be removing the two steps which are concerned with degree skew since
I had not
Github user vasia commented on the pull request:
https://github.com/apache/flink/pull/1671#issuecomment-215494415
But was the difference you saw because of the different logic of the
algorithm or because of using Value types? I was under the impression that the
algorithm was quite
Github user greghogan commented on the pull request:
https://github.com/apache/flink/pull/1671#issuecomment-215474401
@StephanEwen in [FLINK-3768](https://github.com/apache/flink/pull/1896) I
compared an optimal algorithm to the current implementation in Gelly and the
optimal
Github user StephanEwen commented on the pull request:
https://github.com/apache/flink/pull/1671#issuecomment-188918217
Thanks for checking this out.
Do you have an estimate of how this affects Gelly performance for some
algorithms? How much does this improve the performance?
GitHub user greghogan opened a pull request:
https://github.com/apache/flink/pull/1671
[FLINK-3277] Use Value types in Gelly API
You can merge this pull request into a Git repository by running:
$ git pull https://github.com/greghogan/flink
3277_use_value_types_in_gelly_api
