In production, ES runs on a Search Head Cluster (SHC) for high availability, pulling data from an Indexer Cluster for redundancy.
Technology Add-ons (TAs) are deployed to the indexers via the Cluster Master/Manager and to forwarders via the Deployment Server. The ES app itself stays on the SHC.
Data flow: Forwarder → Indexer (with TA for index-time parsing) → Search Head (ES + CIM for search-time enrichment).