Infrastructure Fabric

Responsibilities

• Provide multi-region compute, networking, storage, and routing foundations.
• Define availability zones, redundancy, and disaster recovery posture.
• Expose standardized infrastructure primitives to higher layers.

Key interfaces

• Ingress/egress routing (DNS, edge routing, load balancing).
• Service networking and identity (mTLS/service discovery) where used.
• Storage primitives (object/block/file) and replication contracts.

Operational signals

These are the measurements that tell you whether this layer is healthy in production.

• Regional availability, packet loss, and cross-region latency.
• Capacity headroom and autoscaling events.
• Error budget consumption by region/zone.

Failure modes

• Regional outage or partial zone failure.
• Network partitions and inconsistent routing.
• Capacity depletion during spikes or failover.

Production readiness checklist

• Practice failover (game days) and validate RTO/RPO.
• Use multi-region routing with health-based steering.
• Keep infrastructure changes audited and reversible.