GCP Multi-Region HTTP Load Balancer with Regional Cloud Run Backends

Converted the single-region VM and MIG-based web tier into a structurally different multi-region GCP design using two se...

76 views

Viewed 76 times

Create a free account to remix this architecture

GCP Multi-Region HTTP Load Balancer with Regional Cloud Run Backends

Cloned from an architecture originally created by af9593 View original project

Converted the single-region VM and MIG-based web tier into a structurally different multi-region GCP design using two separate VPCs, regional API gateways, Cloud Run services, Firestore databases, Pub/Sub messaging, and Cloud Storage in us-central1 and us-east1. Added global Cloud DNS for traffic entry and cross-region replication links between data and storage services to improve resilience and regional failover.

intermediate

GCP

Architecture Description

A GCP HTTP load balancer distributed traffic to a single-region managed instance group of Apache web servers, using health checks, proxy, URL map, backend service, and firewall rules to route traffic only to healthy VM instances.

Community Discussion

Be the first to comment

Share your thoughts on this architecture

AI Analysis(1)

AI Review

Tradeoff

about 2 months ago

I see the biggest issue as a mismatch between the stated design and the actual components, and that creates real production risk. The description talks about a global HTTP load balancer sending traffic to a single-region managed instance group of Apache VMs, but the diagram lists regional Cloud Run, API Gateway, Firestore, Pub/Sub, and Storage in two regions with no explicit global load balancer component.

Partha Goswami

@p190188

Hackathon

View portfolio

Try it in the Playground

Open an interactive version — fork it, generate AI variants, or share it with your team.

Open Playground