Cymbal Retail Case Study: Omnichannel Migration Done Right
If EHR Healthcare tests cloud-to-cloud thinking and Mountkirk tests greenfield scale, Cymbal Retail tests the messiest real-world skill: hybrid migration without stopping the business. It's the newest addition to the official case study list, and it catches out candidates who only studied cloud-native greenfield architectures.
The Mission: Modernize Without Stopping the Stores
Cymbal Retail runs physical stores, an e-commerce site, and a large on-premises data center that is aging and capacity-constrained. They want to move analytics and customer-facing services to Google Cloud, but they cannot stop trading while they do it. The exam questions probe whether you can design a phased migration where the legacy estate and the cloud coexist for an extended period.
Exam takeaway: The word "phased" appears strongly in this case study. Answers that require a big-bang cutover are almost always wrong. Look for hybrid connectivity, gradual workload moves, and dual-run periods.
The Architecture Examiners Expect
1. Hybrid Connectivity First
Any Cymbal migration question starts with a link between the legacy data center and GCP. The right layer is Cloud Interconnect (Dedicated or Partner, depending on bandwidth requirements posed) for steady, high-volume paths, or Cloud VPN for lower-volume or interim connectivity. The key exam point is that during a phased migration, systems on both sides must communicate with predictable latency - which leans Interconnect over plain internet paths when the question mentions compliance or guaranteed throughput.
2. Sell-Anywhere Analytics: BigQuery as the Landing Zone
Omnichannel retail means point-of-sale systems, e-commerce events, mobile app analytics, and warehouse management all need to feed one analytical picture. The expected design lands all of them in BigQuery: batch loads from legacy systems, streaming inserts from GCP-native microservices, and direct CSV/Cloud Storage loads. From there, Looker (or Looker Studio) draws the omnichannel dashboards Cymbal's management wants without additional infra.
3. Store Resilience: Point-of-Sale Cannot Stop
The exam's favorite trick for Cymbal Retail is the "cloud link goes down mid-transaction" scenario. The correct answer keeps a local fallback at each store: writes queue locally (often modeled with a lightweight local buffer or a Pub/Sub-emitting edge component) and drain to the cloud when connectivity resumes. A cloud-only POS design that fails closed is the wrong answer - the case study's whole premise is that stores must keep operating.
4. APIs Between Old and New
With a decade of legacy systems, direct integration is not possible for everything. Apigee (or Cloud Endpoints for lighter use cases) exposes inventory, pricing, and customer APIs as managed interfaces so legacy and modern services can consume each other without point-to-point spaghetti.
Frequently Asked Questions (FAQ)
What database should Cymbal Retail use for inventory microservices?
Cloud Spanner if they need strong consistency across regions with relational guarantees; Cloud SQL for simpler regional workloads; Firestore for eventual-consistency catalog data. Match the answer to the consistency language the question uses.
How do they secure customer data during migration?
Encryption at rest and in transit is assumed; VPC Service Controls add a data-exfiltration boundary around BigQuery and data services during the migration window, when data is split between legacy and cloud.
What about Phase 1 of their migration - where does it start?
Analytics moves first. The legacy warehouse is the least risky initial migration target: it's read-heavy, decoupled from transaction paths, and immediately validates GCP as a platform with a visible business win - exactly the kind of phased sequencing exam questions reward.
Does Cymbal Retail ever recommend a full data-center exit?
Only when the question is explicit. Most exam variants ask for hybrid during the transition window, so answers that suddenly declare "everything to cloud in one project" are usually over-reaching and wrong-footed for the phased plan the case study describes.
Summary Checklist
If you see Cymbal Retail, look for these keywords in the correct answer:
- Cloud Interconnect / Cloud VPN for hybrid connectivity
- Phased migration and gradual cutovers (never big-bang)
- BigQuery as the omnichannel analytics landing zone
- Apigee for managed APIs between legacy and cloud
- Local fallback / store-and-forward for point-of-sale resilience
Practice the Cymbal Retail scenario
Our mock exams include Cymbal Retail scenarios alongside all six other official PCA case studies - free, timed, and scored.
Take a Free Mock Exam