All posts tagged

AISIX 1.5.0 extends Model Group routing, retry, failover, balancing, and per-attempt usage records to supported single-request AI endpoints.

AISIX 1.4.0 can serve in degraded mode when Redis is unavailable at startup and can retry usage delivery within strict limits when deduplication evidence permits.

Use Kubernetes 1.37 Memory QoS deliberately to protect API gateway latency, avoid memory cliffs, and roll out cgroup v2 controls safely.

Design an e-commerce gateway around workload isolation, checkout idempotency, safe retries, controlled releases, and peak-traffic failure behavior.