IF-19
Kubernetes horizontal pod autoscaler for warden + dashboard services
📋 Planned
HPA manifests for the warden (FastAPI) and dashboard (Next.js) Deployments. Scales on CPU ≥ 70% and memory ≥ 80%. Min 2 replicas / max 10 per service. Paired with PodDisruptionBudgets (minAvailable: 1) for zero-downtime rolling updates. Metrics served via metrics-server; custom metrics via Prometheus Adapter for P99 latency scaling.
Enterprise
IF-20
Multi-region active-active deployment (EU + US)
📋 Planned
Active-active deployment across EU (Frankfurt) and US (Virginia) regions. Tenant-aware GeoDNS routing, cross-region Postgres streaming replication, Redis active-active via CRDTs, MinIO bucket replication. Automatic failover with < 30s RTO. Sovereign data residency rules enforced at the routing layer.
Enterprise