Failover
Failover
📖 Definition
Failover is a mechanism where the system automatically switches to backup resources when a fault is detected. It ensures that requests can be automatically routed to healthy backup services when the primary service is unavailable, ensuring business continuity.
🔗 How Higress Uses This
Higress supports multi-level Failover: service instance level, service level, and cross-cluster level. AI Gateway particularly supports multi-model Failover, automatically switching to backup models when the primary model fails.
💡 Examples
- 1 Automatically switches to Claude when GPT-4 is unavailable
- 2 Switches to disaster recovery center when primary data center fails
- 3 Automatically removes and routes to healthy instances when service instances fail
🔄 Related Terms
❓ FAQ
What is Failover?
Failover is a mechanism where the system automatically switches to backup resources when a fault is detected. It ensures that requests can be automatically routed to healthy backup services when the primary service is unavailable, ensuring business continuity.
How does Higress support Failover?
Higress supports multi-level Failover: service instance level, service level, and cross-cluster level. AI Gateway particularly supports multi-model Failover, automatically switching to backup models when the primary model fails.
Learn More About Higress
Explore more Higress features and best practices