A global retail firm is migrating its legacy inventory management system to Google Cloud. … — Google Cloud Professional Cloud Architect

A global retail firm is migrating its legacy inventory management system to Google Cloud. The environment consists of a $320\text{ TB}$ archive of unstructured product catalog images and a $750\text{ GB}$ operational PostgreSQL database. The on-premises facility has a shared $100\text{ Mbps}$ internet uplink that must remain available for active business operations. The migration must complete within a target window of 14 days with minimal database downtime during cutover. Which migration strategy should the Cloud Architect recommend?

Order a Google Cloud Transfer Appliance to perform an offline data transfer for the $320\text{ TB}$ image archive to Cloud Storage, and use Database Migration Service (DMS) with continuous Change Data Capture (CDC) over a secure VPN for the PostgreSQL database.Answer
B
Configure Storage Transfer Service to stream the $320\text{ TB}$ image archive directly over the existing $100\text{ Mbps}$ internet link, and use Database Migration Service (DMS) for the PostgreSQL database.
C
Order a Google Cloud Transfer Appliance for the $320\text{ TB}$ image archive, and migrate the operational PostgreSQL database to a Cloud Spanner instance using automated schema conversion and continuous replication.
D
Provision a 10 Gbps Dedicated Interconnect connection to stream the $320\text{ TB}$ image archive via Storage Transfer Service, and perform a one-time database backup and restore using pg_dump during cutover.

Answer

Order a Google Cloud Transfer Appliance for the

320\text{ TB}

image archive, and use Database Migration Service (DMS) with continuous Change Data Capture (CDC) over a VPN for the PostgreSQL database.

The solution combining a Google Cloud Transfer Appliance with Database Migration Service (DMS) via CDC addresses both data scale and network limitations. Moving

320\text{ TB}

of data over a shared

100\text{ Mbps}

pipe takes several months online; Transfer Appliance circumvents this by physically shipping the data to GCP within days. Meanwhile, DMS continuously replicates the

750\text{ GB}

PostgreSQL database using CDC, enabling a low-downtime cutover without requiring lengthy circuit provisioning.

Step-by-Step Solution

Calculate data transfer time over available bandwidth for the bulk dataset.

Transferring

320\text{ TB}

(

3.2 \times 10^{14}

bytes) over a

100\text{ Mbps}

link requires approximately

\frac{320 \times 10^{12} \times 8}{100 \times 10^6} \approx 25,600,000\text{ seconds} \approx 296\text{ days}

, making online transfer infeasible.

Large multi-hundred-terabyte datasets over limited network connections mandate offline physical transfer appliances to meet tight time constraints.

Evaluate operational database migration and cutover requirements.

Database Migration Service (DMS) continuously replicates PostgreSQL data via Change Data Capture (CDC), keeping the target Cloud SQL database synchronized until final cutover.

CDC minimizes operational downtime during database cutover to minutes rather than hours required by offline dumps.

Select appropriate GCP services matching resource scale and timeline.

Combine Transfer Appliance for bulk unstructured data and Database Migration Service over a VPN for relational data.

This combination respects network limitations, completes within the 14-day target window, and avoids over-engineering compute or storage architectures.

Key Concept

Designing hybrid data transfer strategies based on bandwidth constraints, dataset size, downtime tolerance, and lead-time requirements.

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics