Github user aljoscha commented on the pull request:
https://github.com/apache/flink/pull/879#issuecomment-117993213
Forget about the performance numbers I posted. When running the same
benchmark multiple times without shutting down the cluster the performance
degenerates from about 700,000 elements per group to 10,000 elements per group.
---
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.
---
