Resilience, HA & BCDR
1. Start with business services
Define RTO and RPO per business capability, not one universal number.
2. Failure domains
Analyze:
- source failure
- network failure
- integration failure
- transformation failure
- target runtime failure
- identity failure
- consumer failure
- regional service disruption
3. Recovery architecture
For every critical workload document:
- detection
- containment
- retry
- replay
- reconciliation
- escalation
- recovery
- business validation
Current Datasphere tenant configuration documentation includes an SAP HANA multi-availability-zone option for supported configurations. Do not treat platform HA as a substitute for application-level recovery design.
4. BCP exercise
Simulate a critical source outage during a reporting window. Prove which reports continue, which degrade, how freshness is communicated, and how recovery is validated.