Case Studies and Cheat Sheet

Index

• Case Studies List

• Cheat Sheet

• Cross-Case Exam Patterns

• Quick Decision Heuristics

Case Studies

• Altostrat Media Case Study

• Cymbal Retail Case Study

• EHR Healthcare Case Study

• KnightMotives Automotive Case Study

Extra cases used in Sample Questions

• Mountkirk Games

• Helicopter Racing League

• TerramEarth

---

Cheat Sheet

General Strategy

• Read business requirements first

• Eliminate answers that :

  • Over-engineer

  • Increase ops unnecessarily

  • Ignore cost or reliability

1. Mountkirk Games

Industry: Gaming

Business Goals

• Global, low‑latency multiplayer gaming

• Support sudden traffic spikes (game launches)

• Reduce operational overhead

• Enable rapid feature development

Key Requirements

• Stateless, scalable backend

• Global availability

• CI/CD and microservices

• Real‑time analytics

Recommended GCP Solutions

• Compute: GKE (primary), Cloud Run for event-driven services

• Networking: Global HTTP(S) Load Balancer, Cloud CDN

• Data: Cloud Spanner (global consistency) or Firestore

• CI/CD: Cloud Build + Artifact Registry

• Monitoring: Cloud Operations Suite

Exam Tips

• Prefer managed + containerized services

• Design for global scale first

• Choose Spanner when strong consistency + global writes are required

2. TerramEarth

Industry: Heavy equipment & IoT

Business Goals

• Predict equipment failure

• Ingest massive sensor data streams

• Support global field operations

• Minimize downtime

Key Requirements

• High‑throughput data ingestion

• Time‑series and analytical workloads

• ML‑based predictions

• Hybrid connectivity

Recommended GCP Solutions

• Ingestion: Pub/Sub

• Processing: Dataflow (stream + batch)

• Storage: BigQuery, Cloud Storage

• ML: Vertex AI

• Hybrid: Cloud VPN or Dedicated Interconnect

Exam Tips

• Pub/Sub + Dataflow + BigQuery is a classic pipeline

• Use Vertex AI, not custom ML infrastructure

• Emphasize decoupled, event‑driven architecture

3. Helicopter Racing League (HRL)

Industry: Sports & Media

Business Goals

• Live telemetry processing

• Real‑time analytics for broadcasts

• Fan engagement via apps

• Global audience

Key Requirements

• Low‑latency streaming

• High availability during live events

• Edge delivery

• Analytics dashboards

Recommended GCP Solutions

• Ingestion: Pub/Sub

• Stream Processing: Dataflow

• Analytics: BigQuery

• Visualization: Looker / Looker Studio

• Delivery: Cloud CDN

Exam Tips

• Optimize for latency over cost

• Use streaming pipelines, not batch

• Separate ingestion, processing, and analytics layers

4. EHR Healthcare

Industry: Healthcare

Business Goals

• Secure electronic health records

• Meet compliance (HIPAA)

• High availability

• Controlled access

Key Requirements

• Data encryption (at rest & in transit)

• Audit logging

• Fine‑grained IAM

• Disaster recovery

Recommended GCP Solutions

• Compute: GKE or Compute Engine

• Storage: Cloud SQL / Spanner (depending on scale)

• Security: IAM, VPC Service Controls, CMEK

• Monitoring: Cloud Audit Logs

• Backup: Automated backups + multi‑region

Exam Tips

• Always prioritize security and compliance

• Use least‑privilege IAM

• Choose managed databases with encryption enabled by default

---

Cross‑Case Exam Patterns to Remember

Architecture Principles

• Prefer managed services over self‑managed

• Design for high availability & scalability

• Use global services when latency matters

• Decouple systems using Pub/Sub

Data Choices (Very Common on Exam)

Use Case	Best Choice
Global relational DB	Cloud Spanner
OLAP analytics	BigQuery
Streaming ingestion	Pub/Sub
Batch + stream processing	Dataflow
Object storage	Cloud Storage

Security Defaults

• IAM with least privilege

• Encryption by default

• Use VPC Service Controls for sensitive data

• Enable Cloud Audit Logs

---

Quick Decision Heuristics (Exam Gold)

• Global + strong consistency? → Cloud Spanner

• Event‑driven scaling? → Pub/Sub + Cloud Run

• Streaming analytics? → Pub/Sub + Dataflow + BigQuery

• Minimal ops? → Serverless / managed services

• Compliance‑heavy? → Security first, cost second