Hi all, I am in the process of merge MapReduce Runner to its feature branch, https://github.com/apache/beam/pull/3705
I would like to call for contributors' help for making it more mature. Here are areas that need help: 1. Feature completion Currently, there are few ValidatesRunners tests excluded, such as gauge/distribution metrics, stateful/splittable pardo, user timers. 2. Performance improvement For examples, https://issues.apache.org/jira/browse/BEAM-2835 3. Production readiness Try run it in Hadoop cluster Thanks and looking forward to the collaboration on MapReduce Runner. -- Pei
