Federated Search for Amazon S3 (FS-S3) queries Parquet, JSON, or CSV data at rest in S3 backed by AWS Glue Data Catalog tables, metered in Data Scan Units (1 DSU = 10 TB scanned). FS-S3 requires an AWS-hosted Splunk Cloud stack with a standalone Search Head or SHC (not single-instance, GCP, Azure, or FedRAMP/IL5), cannot scan S3 Glacier/Deep Archive objects or Requester Pays buckets, and requires the edit_connections, edit_datasets, and edit_federated_providers capabilities for setup.
To monitor overall stack health, sc_admin users rely on the Cloud Monitoring Console (CMC). CMC provides curated dashboards for Data Quality (line-breaking/timestamp issues), Forwarders, Ingest License Usage, Alerts, and Workload / Splunk Virtual Compute (SVC) consumption.
Under SVC (workload-based) pricing, Splunk provisions compute capacity (CPU/memory) across search and indexing tiers rather than limiting solely by daily GB ingested. CMC SVC dashboards leverage _cmc_summary and splunk-svc-search-attribution to pinpoint heavy scheduled searches, concurrent ad-hoc queries, or skipped searches.
# CMC requires sc_admin role
# Key CMC Dashboards:
# - License Usage (Ingest) & Workload / SVC Usage
# - Alerts > Skipped Searches
# - Data Quality (timestamp & line-breaking warnings)
# - Forwarders (active vs. missing UFs/HFs)