Can I assign ML and Master both roles on single/same node?
From license count perspective, if I use 3 master and 3 ML both roles assigned to those 3 master nodes, will that be considered as 3 License count or 6 License count?
How much RAM as a standard should I assign for JVM for master to run smoothly and for ML
I plan to assign 64 GB RAM with 16 cores per node for above 3 nodes.
The reason I am trying to assign both roles to same node is to save on cost because just for Master (8 or 10 GB RAM) I need to spend entire 1 license cost hence want to share that cost with other licensed role.
Below 6 license are considered separately:
I do plan for HOT (3 nodes) & WARM (3 nodes) again with all 64 GB RAM + 12 core
ok thanks, I did not understood your statement about "its 2 instances" part. There will be actually 3 nodes having dual roles of Master + ML on all 3 nodes.
ok thanks. Its not only elasticsearch but those nodes will also work as ML job processing server. So even if Master (Elasticsearch) & ML server role hosted on same server will be considered as single license you mean?
In such dual role, how should we plan for allocation of available RAM for both roles?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.