This new how-to contains important info about how to use role groups and host templates to easily expand an existing CDH cluster onto heterogeneous hardware:
Thanks - that is helpful info.
First, has any of these changed now that we are in 5.8? It seems the UI is different but the procedure remains the same, but want to verify.
Second, what about a more involved (and perhaps more realistic) scenario? Say a production cluster running multiple services (e.g. Hive, Spark, HBase, Solr), now adding new nodes of different CPU, memory, and disk configs? In particular, Cloudera Manager wizard calculate customized defaults for the many, many memory/resource allocation parameters across all the services based on available #cores and RAM. We want to still keep this feature when we use Host Template. Can we then specify the new role groups partially?