I understand that the HDCloud on AWS comes with 2/3 cluster types. What are the difference services available in each of the cluster types? In the documentation, its stated that for example, "Data Science" cluster type comes with Spark and Zeppelin. Obviously HDFS is part of it; Another example is the EDW-ETL cluster type where it comes with Hive. This means that Tez would also be part of it.
Likewise, where do i get the complete list of services along with the version on each of the cluster types?
My next question is: Is it possible to add more services using Ambari or other means on top of the Services that already comes as part of the cluster type.
Services included are HDFS, YARN, MapReduce2, Tez, Zookeeper, Ambari Metrics, and - if enabled - SmartSense.
Plus, depending on your chosen configuration, you get Hive, Spark, and Zeppelin.
It is not possible to add additional services.
Hope this helps. It's a good point that we should make it clearer in the documentation, I will make sure to do that.
@learninghuman Also, another correction. I checked and the following services are always available:
HDFS, YARN, MapReduce2, Tez, ZooKeeper, Ambari Metrics, Pig, Hive, Spark and - if enabled - SmartSense. In my previous comment, I forgot to mention Pig.
Spark and Hive combination depends on the configuration and Zeppelin is included in the data science and analytics config. See screenshots for details. These are from the latest TP.
1. You can add new service to a deployed cluster on Ambari UI wiht the 'add service' button.
2. Add custom service to cluster by automated way with 'Node Recipes' which is currently in Technical Preview.
As Richard indicates, functionally since HDCloud includes Ambari, it is possible to add services via Ambari.
However, note that you are still subject to the operating constraints - ephemeral clusters and fixed cluster topology (one master node, one worker host group), which is why this is not a documented scenario at this point.
What services are you looking to enable?