Your organization's data governance policy mandates that observability data — including metrics, traces, and logs — must not be retained beyond 90 days. As the Instana self-hosted administrator, which approach correctly controls the data retention period for collected monitoring data?

In a self-hosted Instana deployment, data retention is controlled at the backend level, where administrators configure retention parameters for underlying data stores such as Cassandra for metrics and Elasticsearch for events and spans. These settings are defined in the Instana backend installation configuration and determine how long data persists before being purged automatically. There is no UI-based retention control that operates at the host tag, agent group, or Smart Alert level. The correct answer is 'Adjust the retention period configuration in the Instana backend installation settings, including the data store retention parameters for underlying stores such as Cassandra and Elasticsearch'. The incorrect options — 'Set a 90-day expiry tag on each monitored host in the Instana infrastructure catalog', 'Configure a Smart Alert that triggers a data deletion workflow on a scheduled 90-day basis', 'Enable the Auto-Purge toggle for each agent group under Settings > Agent Management' — are wrong because they do not align with IBM Instana's architecture or recommended practices for this scenario. Understanding this concept is essential for the Domain 7: Security/Compliance domain of the IBM Instana Observability certification.