GCP Cloud Architect case studies
GCP Cloud Architect · Put your knowledge to work in context.
The Google Cloud Professional Cloud Architect credential is aimed at practitioners who can use Google Cloud technologies to design, develop and manage solutions that are robust, secure, scalable, efficient, cost-effective, highly available and flexible - and that drive business objectives. Google frames the role around enterprise cloud strategy, solution design, workload migration approaches, deployment and orchestration, optimization and architectural best practices, and describes familiarity with the Google Cloud Well-Architected Framework as a key requirement, with its pillars (operational excellence, security, reliability, performance optimization, cost optimization and sustainability) woven through the exam objectives. The GCP Cloud Architect case studies are 5 cases and about 6,172 words.
1NorthPeak Retail: latency, residency, and the cost of active-active
NorthPeak Retail runs a pricing and inventory API for 1,200 stores. The service is 85% reads, peaks at 14,000 QPS and 1,900 writes/s, and holds 3.2 TB in Cloud SQL for PostgreSQL.
1,178 words, about 6 min2Halberd Payments: actually reducing PCI scope in a shared VPC estate
Halberd Payments is a PCI DSS Level 1 acquirer running 210 Compute Engine VMs across 44 projects inside a shared VPC host project. The cardholder data environment is 9 of those projects.
1,289 words, about 6 min3Vantage Logistics: unwinding a two-year Terraform drift problem in two sprints
Vantage Logistics has 600 GCP resources managed by 14 Terraform root modules writing into a single state file, applied from engineers' laptops. Three environments share the same set of projects.
1,263 words, about 6 min4Quorum Health: BigQuery cost model before commitment
Quorum Health runs BigQuery on-demand in a single project. The events table is 2.8 PB, unpartitioned, and every dashboard filters on event date .
1,289 words, about 6 min5Trellis Media: an SLO conversation that is really a release conversation
Trellis Media serves a video metadata API behind a global external Application Load Balancer with backends in us-east1 and europe-west1. The SLO is 99.9% availability measured at the load balancer over a rolling 28-day window, which is roughly 40 minutes of error budget.
1,153 words, about 6 min