Hello all - we have a bunch of elasticsearch Nodes running (5 Master/2 Coordinating). Was looking at the community's advise on the current configuration. I feel that JVM Settings may be less than optimal - any comments/suggestions from veterans by seeing this snapshot of a master node will help
Are your instances of ES all running on their own machine comparable to the output you just posted? If that case, since you have 65GB of RAM available, you should probably bump up your HEAP size for the ES JVM. Play around with different values (not exceeding 32G).
For some reference: check here
The rest of the image doesn't say much about how your ES is configured
Hope this helps
thats correct. I have one Ec2 Instance per node. So I do have plenty of RAM to play with. I assume 30 gigs of Heap should be optimal enough in order to use compressed pointers on an RHEL Linux Distro running Open JDK1.8?