Metrics to monitor in ElasticSearch in ELK Deployment

In an ELK deployment what are metrics people typically monitor in Elastic Search. We are able to gather and monitor the following metrics using enter link description heredatadog.

We have a 4nodes in the ELK cluster:-
3 master-data nodes
1 client node + kibana

We have about 2 110 GB indices. We are keeping 7 days logs. So a total of 14 indices.

Please point me to any doc or wiki about ElasticSearch monitoring ( if possible more specific to ElasticSearch-Logstash-Kibana deployment).

Take a look at https://www.elastic.co/guide/en/elasticsearch/guide/current/cluster-admin.html

Thanks Mark for the pointer.