I'm super up for Hadoop 3. One other direction we can go for is to integrate with ML life-cycle tools so that the bigdata -> ML experiments -> production models loop is completed as a whole. One option is MLflow. This is similar to KubeFlow discussion brought up by Jay however we see no feasible plan to integration with Kubernetes and then KubeFlow. MLflow is much lightweight and probably more easier to integrate.
I'll spend sometime to further understand MLflow first. - Evans Ganesh Raju <ganesh.r...@linaro.org> 於 2019年7月18日 週四 上午4:09寫道: > +1 for Distribution based hadoop 3 > I would also prefer inclusion of Apache Ambari mpack. > > Thanks, > Ganesh > > > On Wed, Jul 17, 2019 at 2:09 AM Youngwoo Kim (김영우) <yw...@apache.org> > wrote: > > > Hi folks, > > > > After 1.4.0 release, there is no discussion for the next release yet. so > I > > believe we need to share the ideas and prioritize the items for > > development. > > > > And also https://issues.apache.org/jira/browse/BIGTOP-3123 and > > > > > https://docs.google.com/document/d/1F2Gxu8GARQDZXgqHn12LKkQ5wCV_AF4b_tVmjYB6YfA/edit > > are good starting point for this discussion. > > > > My personal preferences are: > > - Distribution based on Hadoop 3 > > - Up-to-date BigPetStore > > - Software stacks and framework for streaming data & Machine Learning > > - Containers and Cloud Native: What? How? > > > > It would be great to hear your thoughts. > > > > Thanks, > > Youngwoo > > > > > -- > IRC: ganeshraju@#linaro on irc.freenode.ne <http://irc.freenode.net/>t >