Q: You use an AWS Global Accelerator for your application. The backend is an ALB in us-east-1. How does Global Accelerator make the connection faster for a user in Australia compared to pointing their DNS directly to the ALB?
If the Australian user connects directly to the ALB, their TCP handshake and HTTP data traverse the public internet across multiple ISP hops, undersea cables, and peering points—which is slow, jittery, and packet-loss prone.
Want to master this scenario in a live sandbox? KodeKloud's Istio Service Mesh & Advanced Kubernetes Networking Course covers this exact problem with hands-on terminal drills.
🛠️ Production Runbook & Step-by-Step Resolution
Direct Answer: "In our multi-VPC setup, services in private subnets ran into this exact routing obstacle. The interviewer is testing: AWS network backbone, BGP Anycast, TCP termination at the edge.. I structure my answer around systematic triage first, root cause analysis second, and permanent remediation third."
Initial Diagnostics & Root Cause Analysis
If the Australian user connects directly to the ALB, their TCP handshake and HTTP data traverse the public internet across multiple ISP hops, undersea cables, and peering points—which is slow, jittery, and packet-loss prone.
- It uses Anycast IPs, so the Australian user's traffic is immediately routed to the closest AWS Edge Location in Sydney.
- TCP Termination at the Edge: The TCP handshake completes instantly with the Sydney edge node, saving hundreds of milliseconds.
- AWS Backbone: The data then travels from Sydney to us-east-1 strictly over AWS's dedicated, highly optimized, private fiber backbone, bypassing public internet congestion entirely.
Remediation & Permanent Safeguards
With Global Accelerator:
- It uses Anycast IPs, so the Australian user's traffic is immediately routed to the closest AWS Ed...
- TCP Termination at the Edge: The TCP handshake completes instantly with the Sydney edge node, sav...
- AWS Backbone: The data then travels from Sydney to us-east-1 strictly over AWS's dedicated, highl...
Senior hiring managers and bar raisers evaluate how you prioritize during outages. Avoid these common anti-patterns:
- Confusing ALB 502 Bad Gateway (target connection reset) with 504 Gateway Timeout (upstream response timeout) when configuring Networking Global Accelerator 502 indicates the backend target closed the TCP connection prematurely or returned malformed HTTP, whereas 504 means the target failed to respond before idle timeout.
- Setting target deregistration delay to zero instead of allowing connection draining when configuring Networking Global Accelerator Instant target deregistration severs in-flight HTTP requests during auto-scaling scale-in or rolling deployments, causing immediate user-facing drops.
-
Failing to query CloudWatch ELB metrics (
HTTPCode_ELB_5XXvsHTTPCode_Target_5XX) during load balancer outages when configuring Networking Global Accelerator Comparing load balancer generated errors against target generated errors immediately isolates whether the issue lies in the proxy tier or application runtime. - Overlooking Security Group rules on backend EC2/ECS instances that block health check traffic from the ALB when configuring Networking Global Accelerator Target instances must explicitly permit inbound HTTP/HTTPS traffic from the ALB security group on the configured container/application port.