@drcrallen hmm, maybe there's some misunderstanding. As you said, this is still an experimental feature, and exact numbers are not much important at this point. However, we need to add a benchmark implementation which we can do some performance tests, so that we can have better understanding of the implementation of parallel merge algorithm in this PR and how to further improve it. Does it make sense?
[ Full content available at: https://github.com/apache/incubator-druid/pull/5913 ] This message was relayed via gitbox.apache.org for [email protected]
