Github user aalexandrov commented on the pull request: https://github.com/apache/flink/pull/1517#issuecomment-202868272 Overall it seems that the hash-based combiner works better than the sort-based one for (a) uniform, or normal key distribution, and (b) fixed-length records. For skewed key distribution (like Zipf) the two strategies are practically equal, and for variable-length record the extra effort in compacting the record offsets the advanges of the hash-based aggregation approach.
--- 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. ---