Glossary
Updated
Involves architectural patterns (redundant components, graceful degradation, checkpointing, and transaction rollbacks) that ensure AI services remain available and safe under partial failures. Governance prescribes fault-injection testing (chaos engineering), failure-mode analyzes, and clear service-level objectives for recovery time and service continuity.
Example
A cloud-based image-classification API runs on multiple container instances behind a load balancer. If one instance crashes during a high-volume event, traffic automatically shifts to healthy pods, and crashed instances restart without user impact. The ops team regularly performs chaos tests to validate fault-tolerance mechanisms.