Prepare and test your skills
Prepare and test your skills
Free sample question, answered and explained below. The Professional Cloud Security Engineer practice bank has 1,343 more like it. Try 10 free: instant score, no account.
An enterprise is building an automated Dataflow pipeline to sanitize structured customer support CSV files before exporting them to BigQuery for analytics. The files contain fixed fields: Customer_ID, SSN, Age, Signup_Date, and a free-text Agent_Notes field.
The security team has established the following requirements:
SSN by masking all but the last 4 characters.Age into 10-year intervals (e.g., 20-30, 30-40) without incurring content inspection overhead.Signup_Date by a randomized offset to preserve relative time intervals while protecting exact dates.Agent_Notes column with their respective infoType labels.Which Sensitive Data Protection (SDP) configuration should the security team implement?
This configuration attempts to use an unstructured DeidentifyTemplate with ImageTransformations for bounding box redaction, while hardcoding the transformation JSON configuration within the pipeline application code.
recordTransformations, which are required to apply bucketing, field masking, and date shifting to structured CSV columns.ImageTransformations apply specifically to pixel bounding box redactions in image files, not tabular text records.The pipeline was strictly processing image files or scanned document attachments where PII needed to be overwritten with visual black boxes.
A structured Sensitive Data Protection (SDP) DeidentifyTemplate provides a centralized, reusable, and IAM-governed configuration resource that uses recordTransformations to transform structured tabular datasets (such as CSV files or database tables) at the column or field level.
characterMaskConfig applies a non-cryptographic character mask to the SSN column, masking all but the specified number of trailing characters without requiring an inspection pass.fixedSizeBucketingConfig places integer values into predefined discrete ranges (e.g., bucketSize: 10) without content inspection, reducing uniqueness and minimizing API processing overhead.dateShiftConfig shifts dates within a defined lower- and upper-bound range per individual or context, preserving relative time spacing for analytics while obscuring the true date.infoTypeTransformations (with replaceWithInfoTypeConfig) specifically to the Agent_Notes field enables dynamic pattern recognition and replacement of detected PII (such as EMAIL_ADDRESS and PHONE_NUMBER) with token labels.inspectConfig evaluation stage.Agent_Notes).This approach aligns with Google Cloud architectural best practices by isolating resource-heavy infoType inspections to unknown/unstructured fields while executing deterministic, non-cryptographic transformations directly against known structured columns via reusable templates.
This configuration uses BigQuery remote functions backed by a Cloud Run service to invoke the DLP API at query time using cryptographic tokenization methods such as Deterministic Encryption (DE) and Format-Preserving Encryption (FPE-FFX).
The organization required two-way tokenization allowing authorized analysts to re-identify records using a Cloud KMS key directly within BigQuery SQL queries.
This approach relies on an InspectTemplate to execute full-payload inspection across every column in the dataset using built-in detectors (US_SOCIAL_SECURITY_NUMBER, AGE, DATE, EMAIL_ADDRESS), followed by global infoTypeTransformations.
fixedSizeBucketingConfig and dateShiftConfig are primarily implemented as record/field-level transformations on tabular data fields rather than infoType primitive replacements.recordTransformations to bypass unnecessary inspection.The dataset was completely unstructured or had a dynamic, uncharacterized schema where column meanings were unknown prior to ingestion.
Answer them yourself and get your score the moment you finish. No account needed.
Full access includes
Keep the momentum going with these hand-picked practice scenarios
Last updated Validated