You must prove that migrated data matches the source. What is this step?
Why D is the answer
Validation
Validation is the step that proves migrated data matches the source.
In detail
- Validation proves that migrated data matches the source.
- Plan this before you migrate.
Use case
An illustrative scenario, not a real incident.
After a migration, a team compares row counts and checksums, and the source and target match.
See it in different ways
flowchart TD
A{"What are you moving?"} -->|"Huge offline data"| B["Transfer Appliance"]
A -->|"SaaS or warehouse data into BigQuery"| C["BigQuery Data Transfer Service"]
A -->|"A relational database"| D["Database Migration Service"]
A -->|"Ongoing change data capture"| E["Datastream"]stateDiagram-v2 [*] --> Source Source --> TransferAppliance: petabytes, slow network Source --> Datastream: database changes, continuous Source --> DataTransferService: SaaS on a schedule TransferAppliance --> Validate Datastream --> Validate DataTransferService --> Validate Validate --> [*]
flowchart LR S0["Migrate"] --> S1["Compare counts"] --> S2["Compare values"] --> S3["Sign off"]
Key point: Validation proves that data matches.
Why the other options are wrong
- A. Partitioning
- Partitioning organizes tables.
- B. Materialization
- That precomputes results.
- C. Reservation
- That buys capacity.