Certified Data Professional (CDP) — Questions and Answers

Question 1: What is 'amortization' in the context of financial data for intangible assets?

Transferring asset cost to a reserve account
Restating historical cost to current fair market value
Writing off the full cost of an asset in the year of purchase
Systematically expensing the cost of an intangible asset over its useful life (Correct answer)

Correct answer: Systematically expensing the cost of an intangible asset over its useful life

Amortization allocates the cost of intangible assets (e.g., patents, licenses) as an expense over their expected useful economic life.

Question 2: What is the primary purpose of data skew mitigation techniques like 'salting' in distributed big data joins?

To cache frequently joined datasets in memory
To encrypt sensitive join keys before transmission
To convert wide joins into narrow transformations
To distribute data more evenly across partitions and reduce hotspots (Correct answer)

Correct answer: To distribute data more evenly across partitions and reduce hotspots

Salting adds a random prefix to skewed keys to distribute them across multiple partitions, reducing the load on overloaded nodes.

Question 3: In Apache Kafka, what is the role of the 'consumer group'?

To allow multiple consumers to share the work of reading a topic's partitions in parallel (Correct answer)
To replicate topic partitions across multiple brokers for durability
To compress messages across partitions for storage efficiency
To batch-produce messages to a topic from multiple producer threads

Correct answer: To allow multiple consumers to share the work of reading a topic's partitions in parallel

A consumer group allows multiple consumers to each read from a distinct subset of partitions, enabling parallel consumption of a topic.

Question 4: In a data integration ETL process, what is the purpose of a 'lookup' transformation?

Sorting records by a specified key column before loading
Splitting a single incoming source stream into multiple target output streams
Retrieving matching values from a reference table to enrich or validate records being processed (Correct answer)
Removing exact duplicate records from the incoming dataset

Correct answer: Retrieving matching values from a reference table to enrich or validate records being processed

A lookup transformation queries a reference or dimension table to find matching values, enabling records to be enriched with additional attributes or validated against known reference data.

Question 5: What does 'grain' mean in the context of a data warehouse fact table?

The level of detail that each row in the fact table represents (Correct answer)
The type of aggregation applied to measures
The number of dimensions linked to the fact table
The number of rows in the fact table

Correct answer: The level of detail that each row in the fact table represents

The grain of a fact table defines the most atomic level of detail that each row represents, such as one row per individual sales transaction line item.

Question 6: In data interoperability, what does 'semantic interoperability' specifically ensure?

That the meaning and context of exchanged data is consistently and correctly understood across different systems (Correct answer)
That data transfer speeds are optimized to meet service level agreements between systems
That encryption and security standards are mutually compatible between exchanging systems
That network transport protocols are standardized across all systems in the enterprise

Correct answer: That the meaning and context of exchanged data is consistently and correctly understood across different systems

Semantic interoperability ensures that when data is exchanged between systems, both systems interpret the meaning, context, and business definitions of that data in the same consistent way.

Question 7: What is 'principle of least privilege' in database access control?

Granting users only the minimum permissions required to perform their job functions (Correct answer)
Restricting database access to only privileged administrators
Granting permissions for the least-used database objects
Granting all users read access to all tables

Correct answer: Granting users only the minimum permissions required to perform their job functions

The principle of least privilege grants each user or process only the minimum database permissions necessary for their legitimate work, limiting the damage from errors or attacks.

Question 8: Which of the following best describes a Data Quality dimension?

A storage format for data files
A measurable characteristic of data such as accuracy, completeness, or timeliness (Correct answer)
A type of ETL transformation
A database index type

Correct answer: A measurable characteristic of data such as accuracy, completeness, or timeliness

Data Quality dimensions are measurable characteristics like accuracy, completeness, consistency, timeliness, and uniqueness used to assess data quality.

Question 9: What does 'data lineage' mean in the context of data governance?

Classifying data by sensitivity level
Defining data retention periods
Documenting the origin, movement, and transformation of data over time (Correct answer)
Tracking the age of data records

Correct answer: Documenting the origin, movement, and transformation of data over time

Data lineage tracks where data originated, how it has moved through systems, and what transformations it has undergone.

Question 10: What is a key advantage of ELT over traditional ETL in a cloud data warehouse environment?

ELT leverages the target system's massively parallel processing power for transformations (Correct answer)
ELT only supports structured relational data sources
ELT requires significantly less storage space in the cloud
ELT eliminates the need for any data quality checks

Correct answer: ELT leverages the target system's massively parallel processing power for transformations

ELT loads raw data first, then uses the cloud data warehouse's powerful parallel processing to perform transformations, taking advantage of scalable compute resources.

Question 11: In data governance, what is a 'data steward' most likely NOT responsible for?

Maintaining data definitions in the business glossary
Resolving data quality issues
Physically administering database servers (Correct answer)
Defining data quality rules

Correct answer: Physically administering database servers

Physical database server administration is the responsibility of Database Administrators, not Data Stewards, who focus on policy, quality, and definitions.

Question 12: What is denormalization in database design?

The process of converting ER diagrams to physical tables
Removing unused tables from a schema
Intentionally introducing redundancy into a database to improve read performance (Correct answer)
Converting a relational model to a document model

Correct answer: Intentionally introducing redundancy into a database to improve read performance

Denormalization intentionally adds redundant data to a normalized schema to reduce joins and improve query read performance, often used in data warehouses.

Question 13: In portfolio analytics, what does 'alpha' represent?

Excess return of an investment relative to a benchmark index (Correct answer)
The volatility of a portfolio relative to the market
The total absolute return of a portfolio
The correlation between a portfolio and its benchmark

Correct answer: Excess return of an investment relative to a benchmark index

Alpha measures the active return on an investment above or below a benchmark, representing the value added by the portfolio manager.

Question 14: In a snowflake schema, how do dimension tables differ from those in a star schema?

They contain fact data instead of descriptive data
They store historical data using SCD techniques
They are normalized into multiple related tables rather than a single denormalized table (Correct answer)
They do not have primary keys

Correct answer: They are normalized into multiple related tables rather than a single denormalized table

In a snowflake schema, dimension tables are normalized into multiple related sub-tables, reducing redundancy compared to the denormalized dimension tables in a star schema.

Question 15: What is the primary advantage of a Data Vault modeling approach over traditional star/snowflake schemas?

Faster OLAP query performance
Simpler queries for end users
Better auditability, flexibility for change, and support for loading from multiple source systems (Correct answer)
Reduced storage requirements

Correct answer: Better auditability, flexibility for change, and support for loading from multiple source systems

Data Vault modeling excels at auditability, handles source system changes gracefully, and supports parallel loading from multiple sources using its Hub-Satellite-Link structure.

Question 16: In enterprise data architecture, what is a 'Data Domain'?

A database server hostname
A logical grouping of related data assets under a common subject area or business function (Correct answer)
A network partition for data storage
A type of data encryption algorithm

Correct answer: A logical grouping of related data assets under a common subject area or business function

A Data Domain is a logical grouping of related data (e.g., Customer, Product, Finance) organized around a common business subject area or function.

Question 17: Which best reflects the sequence in which data is supplied to a Reducer's reduce function, assuming default settings?

The keys given to a Reducer aren't in a predictable order, but the values associated with those keys always are
Both the keys and values passed to a Reducer always appear in sorted order
Neither keys nor values are in any predictable order
The keys given to a Reducer are in sorted order, but the values associated with each key are in no predictable order (Correct answer)

Correct answer: The keys given to a Reducer are in sorted order, but the values associated with each key are in no predictable order

Although the values attached to each key are delivered to a Reducer in sorted order, there is no consistency in the order of the values.

Question 18: Which characteristic of big data refers to the trustworthiness and accuracy of the data collected?

Veracity (Correct answer)
Variety
Volume
Velocity

Correct answer: Veracity

Veracity refers to the uncertainty, noise, and reliability of data — ensuring data quality and accuracy in big data systems.

Question 19: What does 'idempotency' mean in the design of data integration pipelines?

The ability to process multiple data streams simultaneously without conflicts
The capacity for the pipeline to auto-scale based on incoming data volume
Automatic detection and parsing of schemas from unstructured data sources
Ensuring that executing the same integration process multiple times produces the same result without creating duplicates (Correct answer)

Correct answer: Ensuring that executing the same integration process multiple times produces the same result without creating duplicates

An idempotent pipeline can be safely re-run after failures without creating duplicate records or corrupting data, because repeated execution yields the same outcome as a single execution.

Question 20: In a star schema, what does the 'fact table' contain?

Measurable, quantitative business metrics and foreign keys to dimension tables (Correct answer)
Configuration settings for the data warehouse
User access permissions
Descriptive attributes about business entities

Correct answer: Measurable, quantitative business metrics and foreign keys to dimension tables

The fact table in a star schema contains quantitative business metrics (measures) and foreign keys linking to surrounding dimension tables.

Question 21: What is database normalization's Third Normal Form (3NF) requirement?

No transitive dependencies — non-key attributes must depend only on the primary key, not on other non-key attributes (Correct answer)
No partial dependencies on a composite key
All foreign keys must reference primary keys in other tables
All attributes must be atomic with no repeating groups

Correct answer: No transitive dependencies — non-key attributes must depend only on the primary key, not on other non-key attributes

3NF requires that non-key attributes depend only on the primary key (not on other non-key attributes), eliminating transitive dependencies that cause update anomalies.

Question 22: What is 'data classification' in the context of information security governance?

Archiving old data records
Sorting data alphabetically in a database
Categorizing data by sensitivity level to determine appropriate security controls (Correct answer)
Indexing data for faster retrieval

Correct answer: Categorizing data by sensitivity level to determine appropriate security controls

Data classification categorizes data by sensitivity (e.g., public, internal, confidential, restricted) to apply appropriate security and handling controls.

Question 23: In a logical data model, what is the difference from a physical data model?

Logical models always use star schemas
Logical models include hardware specifications; physical models do not
Logical models define business entities and relationships without platform specifics; physical models include DBMS-specific implementation details (Correct answer)
Logical models are used only by developers; physical models are for business users

Correct answer: Logical models define business entities and relationships without platform specifics; physical models include DBMS-specific implementation details

Logical models define entities, attributes, and relationships in business terms without DBMS specifics, while physical models include tables, data types, indexes, and storage details.

Question 24: What is the purpose of a database transaction log?

To store backup copies of database tables
To record all changes made to the database, enabling recovery and replication (Correct answer)
To track query performance metrics
To store user login history

Correct answer: To record all changes made to the database, enabling recovery and replication

The transaction log records every modification to the database, enabling point-in-time recovery, undo/redo operations, and replication to secondary servers.

Question 25: What is the purpose of a 'reference data' table in a database?

Link two fact tables together
Store large binary files
Provide a controlled set of valid values for data fields, such as country codes or status types (Correct answer)
Archive historical transaction data

Correct answer: Provide a controlled set of valid values for data fields, such as country codes or status types

Reference data tables store stable, controlled sets of valid values (like country codes, currency codes, or status enumerations) used to ensure data consistency.

Question 26: In Service-Oriented Architecture (SOA), which principle is most critical for achieving system interoperability?

Services are tightly coupled directly to their specific consuming applications
Services expose standard, language-agnostic interfaces that any compliant consumer can invoke (Correct answer)
Services are accessible only within the internal corporate network perimeter
All services in the enterprise share a single centralized monolithic database

Correct answer: Services expose standard, language-agnostic interfaces that any compliant consumer can invoke

SOA achieves interoperability by requiring services to use standard, technology-neutral interfaces (such as WSDL/SOAP or REST), allowing any system to consume them regardless of underlying implementation.

Question 27: What is the purpose of a staging area in a data warehouse ETL process?

To serve final reports to end users
To temporarily hold extracted source data before transformation and loading into the warehouse (Correct answer)
To cache frequently-run queries
To store the final dimensional model

Correct answer: To temporarily hold extracted source data before transformation and loading into the warehouse

The staging area is a temporary storage zone where raw extracted data is held and processed before being transformed and loaded into the permanent data warehouse tables.

Question 28: Which integration architecture pattern uses a centralized hub through which all application messages are routed, mediated, and transformed?

Flat file exchange over SFTP
Direct database linking via shared schemas
Enterprise Service Bus (ESB) (Correct answer)
Point-to-point integration

Correct answer: Enterprise Service Bus (ESB)

An Enterprise Service Bus (ESB) provides a central communication backbone that mediates, routes, and transforms messages between all connected applications.

Question 29: Which of the following is a primary output of a data stewardship program?

Reduced storage costs
Improved data quality and trusted data assets (Correct answer)
Faster ETL processing
Faster query execution

Correct answer: Improved data quality and trusted data assets

Data stewardship programs primarily aim to improve data quality and create trusted, authoritative data assets for business use.

Question 30: What role does metadata play in supporting data integration processes?

It encrypts data records during transmission between source and target systems
It generates end-user interfaces for data entry and validation workflows
It stores the actual data values that are being integrated between systems
It describes the structure, origin, lineage, and transformation history of integrated data (Correct answer)

Correct answer: It describes the structure, origin, lineage, and transformation history of integrated data

Metadata provides context about integrated data — including its structure, source systems, transformation rules applied, and lineage — enabling teams to understand, govern, and troubleshoot integrations.

Question 31: What is 'dynamic data masking' and how does it differ from static data masking?

Dynamic masking is used for backups; static masking is for production
Dynamic masking shows masked data in query results to unauthorized users without altering stored data; static masking creates a separate masked copy of the data (Correct answer)
They are identical techniques with different names
Dynamic masking physically alters stored data; static masking shows masked data only to some users

Correct answer: Dynamic masking shows masked data in query results to unauthorized users without altering stored data; static masking creates a separate masked copy of the data

Dynamic data masking obfuscates data in query results in real-time without changing stored data, while static masking creates a separate, permanently masked copy for non-production use.

Question 32: In database recovery, what is the difference between a 'full backup' and a 'differential backup'?

Differential backups include transaction logs; full backups do not
Full backups are faster; differential backups are more complete
A full backup copies all data; a differential backup copies only data changed since the last full backup (Correct answer)
Full backups are stored offsite; differential backups are stored locally

Correct answer: A full backup copies all data; a differential backup copies only data changed since the last full backup

A full backup captures all database data, while a differential backup captures only data that has changed since the most recent full backup, reducing backup time and storage.

Question 33: Which data governance framework component assigns formal accountability for data assets?

Data Masking
Data Profiling
Data Cataloging
Data Ownership (Correct answer)

Correct answer: Data Ownership

Data Ownership assigns formal accountability to individuals or teams for the quality, access, and proper use of specific data assets.

Question 34: Which lightweight, human-readable data interchange format has become widely dominant for REST API data exchange?

XML (Extensible Markup Language)
Fixed-width flat file
COBOL copybook format
JSON (JavaScript Object Notation) (Correct answer)

Correct answer: JSON (JavaScript Object Notation)

JSON has become the dominant format for REST API data exchange due to its lightweight structure, human readability, and native compatibility with modern programming languages.

Question 35: What is a conceptual data model?

A model that defines data types and constraints
A detailed physical design of database tables and indexes
A diagram showing ETL data flows
A high-level representation of business concepts and relationships without technical detail (Correct answer)

Correct answer: A high-level representation of business concepts and relationships without technical detail

A conceptual data model provides a high-level, business-focused representation of key entities and relationships without technical implementation details.

Question 36: Which regulatory framework requires financial institutions to report derivative positions to a trade repository?

Sarbanes-Oxley Act
Basel III
MiFID II
Dodd-Frank Act (CFTC reporting) (Correct answer)

Correct answer: Dodd-Frank Act (CFTC reporting)

The Dodd-Frank Wall Street Reform Act mandated reporting of over-the-counter derivative transactions to CFTC-registered trade repositories.

Question 37: Which type of data governance maturity model stage describes ad hoc, undocumented data management with no formal policies?

Initial/Reactive (Correct answer)
Optimized
Defined
Managed

Correct answer: Initial/Reactive

The Initial or Reactive stage of data governance maturity is characterized by ad hoc, informal, and undocumented data management practices.

Question 38: Why is Machine Learning Used in Data Science?

All the above
For Cleaning
For Visualization
For Prediction (Correct answer)

Correct answer: For Prediction

Machine learning plays a crucial role in data science, particularly in the context of prediction.

Question 39: What is a primary benefit of API-led connectivity in modern enterprise integration architectures?

It promotes reusable integration assets organized into System, Process, and Experience API layers (Correct answer)
It automatically replaces and decommissions all legacy on-premise systems
It restricts integrations exclusively to cloud-native SaaS applications
It completely eliminates the need for any ETL or data transformation processing

Correct answer: It promotes reusable integration assets organized into System, Process, and Experience API layers

API-led connectivity organizes APIs into three reusable layers — System (unlock data), Process (orchestrate), and Experience (deliver) — enabling flexible, composable integrations.

Question 40: In Hadoop MapReduce, what happens during the 'shuffle and sort' phase?

Combiners aggregate all intermediate results before shuffle begins
The final output is written to HDFS in sorted order
Input data is split into fixed-size blocks and distributed to mappers
Mapper output is transferred to reducers and sorted by key before reduction (Correct answer)

Correct answer: Mapper output is transferred to reducers and sorted by key before reduction

During shuffle and sort, the framework transfers mapper outputs to the appropriate reducers and sorts them by key so each reducer receives all values for a given key.

Question 41: Which role is primarily responsible for defining data policies and standards within an organization?

ETL Developer
Database Administrator
Data Analyst
Data Steward (Correct answer)

Correct answer: Data Steward

Data Stewards are responsible for defining and enforcing data policies, standards, and procedures across the organization.

Question 42: What is the primary role of a message broker in a data integration architecture?

Generating business intelligence reports from integrated data
Permanently storing integrated data for analytics
Routing messages between producers and consumers while decoupling them (Correct answer)
Encrypting all data in transit between systems

Correct answer: Routing messages between producers and consumers while decoupling them

A message broker decouples producers and consumers by receiving messages from producers and routing them to the appropriate consumers, enabling asynchronous communication.

Question 43: What is a Slowly Changing Dimension (SCD) Type 2?

A dimension that keeps only the original value forever
A dimension that tracks history by adding new rows with effective date ranges when attributes change (Correct answer)
A dimension where old values are overwritten with new values
A dimension with no historical tracking

Correct answer: A dimension that tracks history by adding new rows with effective date ranges when attributes change

SCD Type 2 tracks historical changes by adding new rows with start/end dates, allowing analysis of both current and historical dimension attribute values.

Question 44: What does ETL stand for in the context of data integration?

Extract, Transform, Load (Correct answer)
Encode, Transform, Load
Extract, Transfer, Load
Extract, Transfer, Link

Correct answer: Extract, Transform, Load

ETL stands for Extract, Transform, Load — the core three-step process of moving data from source systems to a target system.

Question 45: What is the role of the HDFS Checkpoint Node (formerly Secondary NameNode)?

Storing encrypted keys for HDFS Transparent Encryption
Serving as a hot standby that takes over if the primary NameNode fails
Periodically merging the fsimage and edit log to create a new checkpoint (Correct answer)
Balancing read requests across multiple NameNodes

Correct answer: Periodically merging the fsimage and edit log to create a new checkpoint

The Checkpoint Node (Secondary NameNode) regularly combines the current fsimage and accumulated edits into a new fsimage, preventing the edit log from growing indefinitely.

Question 46: Which architectural pattern separates compute from storage to allow independent scaling in cloud data platforms?

Lambda Architecture
Shared-Nothing Architecture
Federated Architecture
Decoupled Storage-Compute Architecture (Correct answer)

Correct answer: Decoupled Storage-Compute Architecture

Decoupled Storage-Compute Architecture allows storage (e.g., S3) and compute (e.g., query engines) to scale independently, reducing costs and improving flexibility.

Question 47: What distinguishes OLTP systems from OLAP systems in terms of design purpose?

OLTP supports historical data; OLAP supports only current data
OLTP uses columnar storage; OLAP uses row-based storage
OLTP is optimized for high-volume, short transactions; OLAP is optimized for complex analytical queries (Correct answer)
OLTP is for complex analytics; OLAP is for simple transactions

Correct answer: OLTP is optimized for high-volume, short transactions; OLAP is optimized for complex analytical queries

OLTP systems are designed for high-volume, short-duration transactions (insert/update/delete), while OLAP systems are designed for complex, read-heavy analytical queries.

Question 48: Which technique in big data preprocessing addresses class imbalance in training datasets?

Oversampling minority classes or undersampling majority classes (e.g., SMOTE) (Correct answer)
Normalizing feature values to a 0–1 range
Dimensionality reduction using PCA
Removing duplicate records from the training set

Correct answer: Oversampling minority classes or undersampling majority classes (e.g., SMOTE)

Techniques like SMOTE (Synthetic Minority Oversampling Technique) generate synthetic samples for minority classes to balance the training distribution.

Question 49: Which data modeling technique uses entities, attributes, and relationships to represent business concepts?

Object-Oriented Modeling
Entity-Relationship (ER) Modeling (Correct answer)
Physical Data Modeling
Dimensional Modeling

Correct answer: Entity-Relationship (ER) Modeling

Entity-Relationship modeling represents data as entities (business objects), attributes (properties), and relationships (associations) between entities.

Question 50: Which normal form requires that every non-key attribute be fully dependent on the entire primary key?

Third Normal Form (3NF)
Boyce-Codd Normal Form (BCNF)
Second Normal Form (2NF) (Correct answer)
First Normal Form (1NF)

Correct answer: Second Normal Form (2NF)

Second Normal Form (2NF) requires that a table be in 1NF and that every non-key attribute be fully functionally dependent on the entire composite primary key.

Question 51: Which data integration approach provides a virtual unified view of data across multiple sources without physically moving or copying the data?

Data federation (Correct answer)
Data replication
Data migration
Data consolidation

Correct answer: Data federation

Data federation creates a virtual unified view by querying source systems on demand, leaving the data in place rather than copying it to a central repository.

Question 52: What is the purpose of 'checkpointing' in distributed stream processing frameworks like Flink or Spark Streaming?

To validate that output records match expected values during integration testing
To throttle input data rates when downstream consumers cannot keep pace
To periodically snapshot the state of a streaming job so it can be recovered after a failure (Correct answer)
To compress accumulated stream state to reduce memory consumption during long-running jobs

Correct answer: To periodically snapshot the state of a streaming job so it can be recovered after a failure

Checkpointing saves a consistent snapshot of operator state and stream positions, allowing the job to resume from the last checkpoint after a failure rather than restarting from scratch.

Question 53: Which messaging standard is most commonly used to ensure interoperability between healthcare information systems?

HL7 (Correct answer)
FIX Protocol
ISO 20022
SWIFT MT messages

Correct answer: HL7

HL7 (Health Level 7) is the dominant standard for exchanging clinical and administrative data between healthcare applications and systems.

Question 54: In HDFS, what is a 'rack-aware' block placement policy designed to achieve?

Reducing NameNode metadata size by grouping blocks per rack
Maximizing write throughput by placing all replicas on the same DataNode
Ensuring blocks are always placed on the DataNode closest to the client
Balancing fault tolerance and bandwidth by spreading replicas across racks (Correct answer)

Correct answer: Balancing fault tolerance and bandwidth by spreading replicas across racks

Rack-aware placement puts one replica locally, one on a different rack node, and one more on that second rack so a single rack failure doesn't lose data.

Question 55: How can a client access a file on HDFS?

The client simultaneously requests every DataNode. The client receives a response from the DataNode with the requested data. Data is immediately read by the client from the DataNode.
For the block location, the Client contacts the NameNode (s). The NameNode contacts the DataNode that contains the desired data block. Data is exchanged from the DataNode to the NameNode to the Client from the NameNode to the Client.
The client asks NameNode where the block is located (s). The NameNode gives the client the block location(s) back. Data is immediately read by the client from the DataNode (s)
For the block location, the client contacts the NameNode (s). The NameNode then requests block locations from the DataNodes. The NameNode leads the client to the DataNode, holding the requested data block when the DataNodes react to it (s). Data is then read directly from the DataNode by the client. (Correct answer)

Correct answer: For the block location, the client contacts the NameNode (s). The NameNode then requests block locations from the DataNodes. The NameNode leads the client to the DataNode, holding the requested data block when the DataNodes react to it (s). Data is then read directly from the DataNode by the client.

For the block location, the client contacts the NameNode (s). The NameNode then requests block locations from the DataNodes. The NameNode leads the client to the DataNode, holding the requested data block when the DataNodes react to it (s). The data is then immediately read by the client from the DataNode.

Question 56: In Apache Spark, what is the significance of the 'DAG' (Directed Acyclic Graph) scheduler?

It stores RDD lineage information as a serialized graph in HDFS for checkpointing
It routes shuffle data between executors using a graph-based routing protocol
It manages cluster resource allocation by tracking node CPU and memory availability
It represents the logical execution plan of a Spark job, enabling optimization and fault recovery (Correct answer)

Correct answer: It represents the logical execution plan of a Spark job, enabling optimization and fault recovery

Spark's DAG scheduler translates user operations into a graph of stages and tasks, optimizing the execution plan and enabling recomputation of lost partitions from lineage.

Question 57: What is the purpose of normalization in relational database design?

Create indexes for faster retrieval
Increase storage capacity
Eliminate data redundancy and ensure data integrity by organizing data into related tables (Correct answer)
Improve query performance by adding redundancy

Correct answer: Eliminate data redundancy and ensure data integrity by organizing data into related tables

Normalization organizes data into related tables to eliminate redundancy, reduce anomalies, and ensure data integrity in relational databases.

Question 58: During a typical sort and shuffle phase of MapReduce, how are keys and data presented and delivered to Reducers?

Keys are presented to a Reducer in random order; values for a given key are sorted in ascending order
Keys are presented to a Reducer in random order; values for a given key are not sorted
Keys are presented to a Reducer in sorted order; values for a given key are sorted in ascending orde
Keys are presented to a Reducer in sorted order; values for a given key are not sorted (Correct answer)

Correct answer: Keys are presented to a Reducer in sorted order; values for a given key are not sorted

Values for a particular key are not sorted, although keys are delivered to reducers in sorted order.

Question 59: What is a data catalog's primary function in a modern data architecture?

Execute ETL jobs and data transformations
Manage database user permissions
Store and process raw data at scale
Provide a searchable inventory of data assets with metadata, lineage, and business context (Correct answer)

Correct answer: Provide a searchable inventory of data assets with metadata, lineage, and business context

A data catalog acts as an organized inventory of data assets, providing metadata, lineage, and business context to help users discover and understand available data.

Question 60: What is a Data Governance Council primarily responsible for?

Managing database server hardware
Performing daily data backups
Writing ETL scripts
Overseeing enterprise data governance policies and resolving data conflicts (Correct answer)

Correct answer: Overseeing enterprise data governance policies and resolving data conflicts

A Data Governance Council oversees the implementation of data governance policies and arbitrates disputes over data ownership and definitions.

Question 61: Under GDPR, what is the maximum time an organization typically has to report a data breach to authorities?

7 days
48 hours
24 hours
72 hours (Correct answer)

Correct answer: 72 hours

GDPR requires organizations to notify supervisory authorities of a personal data breach within 72 hours of becoming aware of it.

Question 62: What is the role of metadata in data governance?

It stores the actual business data
It provides context and information about data assets to aid in discovery and management (Correct answer)
It performs data transformations
It replaces the need for data documentation

Correct answer: It provides context and information about data assets to aid in discovery and management

Metadata provides context about data assets—such as structure, origin, and usage—enabling better discovery, understanding, and management of data.

Question 63: In financial data quality management, what does 'referential integrity' ensure for trade data linked to instrument master data?

Counterparty names are spelled consistently across records
All trade amounts are within predefined thresholds
Every trade record references a valid, existing instrument in the security master (Correct answer)
Trade timestamps are consistent across all source systems

Correct answer: Every trade record references a valid, existing instrument in the security master

Referential integrity enforces that foreign key values in trade records correspond to actual records in the referenced security master table.

Question 64: Which data integration pattern is best suited for near-real-time data movement requirements?

Monthly batch file transfers via SFTP
Event-driven streaming integration (Correct answer)
Quarterly data migration projects
Manual CSV imports by data stewards

Correct answer: Event-driven streaming integration

Event-driven streaming integration processes and delivers data as events occur, making it ideal for near-real-time requirements where low latency is critical.

Question 65: Which data architecture style uses a central repository where raw data is stored in its native format until needed?

Operational Data Store (ODS)
Data Vault
Data Mart
Data Lake (Correct answer)

Correct answer: Data Lake

A Data Lake stores raw data in its native format at scale, allowing flexible analysis later, unlike data warehouses that require structure upfront.

Question 66: What does 'right to erasure' (also called 'right to be forgotten') under GDPR mean?

Employees can delete business records
Data must be erased after 5 years
Organizations can delete any data at will
Individuals can request that their personal data be deleted under certain conditions (Correct answer)

Correct answer: Individuals can request that their personal data be deleted under certain conditions

Under GDPR, individuals have the right to request deletion of their personal data when it's no longer necessary for the purpose it was collected.

Question 67: What is the purpose of a 'wash sale rule' in financial data compliance?

Require immediate settlement of all equity trades
Mandate daily reconciliation of trading accounts
Prevent double-counting of dividend income
Disallow a tax loss deduction if the same security is repurchased within 30 days (Correct answer)

Correct answer: Disallow a tax loss deduction if the same security is repurchased within 30 days

The IRS wash sale rule disallows claiming a capital loss if a substantially identical security is bought within 30 days before or after the sale.

Question 68: What is Change Data Capture (CDC) in the context of data integration?

A technique that identifies and captures only the data that has changed since the last extraction (Correct answer)
A process for archiving full historical snapshots on a schedule
A method for encrypting sensitive data columns before transfer
A governance policy framework for documenting data changes

Correct answer: A technique that identifies and captures only the data that has changed since the last extraction

CDC tracks modifications (inserts, updates, deletes) in source systems and captures only those changed records, enabling efficient incremental data integration.

Question 69: Which file format is optimized for columnar storage and is widely used in big data ecosystems for analytical queries?

Avro
CSV
JSON
Parquet (Correct answer)

Correct answer: Parquet

Parquet is a columnar storage format that enables efficient compression and encoding schemes, significantly improving analytical query performance.

Question 70: Which regulation specifically governs how healthcare organizations in the US must handle patient data?

SOX
GDPR
CCPA
HIPAA (Correct answer)

Correct answer: HIPAA

HIPAA (Health Insurance Portability and Accountability Act) sets US federal standards for protecting sensitive patient health information.

Question 71: What is a surrogate key in database design?

A system-generated, meaningless unique identifier assigned to each row in a table (Correct answer)
A composite key made of multiple columns
A foreign key referencing another table
A natural business key like a Social Security Number

Correct answer: A system-generated, meaningless unique identifier assigned to each row in a table

A surrogate key is a system-generated, meaningless numeric identifier (like an auto-increment ID) used as a primary key independent of business data.

Question 72: Which consensus algorithm is commonly used by distributed systems like ZooKeeper to achieve leader election and coordination?

Two-Phase Commit
Paxos (Correct answer)
Gossip Protocol
Raft

Correct answer: Paxos

ZooKeeper uses the ZAB (ZooKeeper Atomic Broadcast) protocol, which is closely based on the Paxos consensus algorithm.

Question 73: What primary challenge does data heterogeneity create in enterprise integration projects?

It dramatically increases the physical storage requirements of the target data warehouse
It legally prevents data from being transferred across different geographic regions
It makes applying encryption to data impossible during the transfer process
Different data formats, schemas, and semantic meanings across sources require complex reconciliation work (Correct answer)

Correct answer: Different data formats, schemas, and semantic meanings across sources require complex reconciliation work

Heterogeneous sources differ in file formats, data types, naming conventions, and the business meaning of fields, requiring significant effort to reconcile and produce consistent integrated data.

Question 74: What is 'role-based access control' (RBAC) in database security?

Granting permissions to individual users directly
Restricting access based on the time of day
Encrypting data based on a user's department
Assigning permissions to roles and then assigning users to appropriate roles to manage access efficiently (Correct answer)

Correct answer: Assigning permissions to roles and then assigning users to appropriate roles to manage access efficiently

RBAC assigns permissions to roles (e.g., read-only analyst, data editor) and then assigns users to those roles, simplifying access management at scale.

Question 75: What is the purpose of a staging area in a data integration architecture?

To enforce role-based access control policies on the integrated data assets
To generate analytical dashboards and reports directly for business end users
To provide a temporary zone where data is validated, cleansed, and transformed before loading to the target (Correct answer)
To serve as the permanent operational storage location for all integrated data

Correct answer: To provide a temporary zone where data is validated, cleansed, and transformed before loading to the target

A staging area is a transient workspace that holds data extracted from source systems while it undergoes validation, cleansing, and transformation before being loaded into the final target.

Question 76: What is the CAP theorem in distributed database systems?

A theorem about maximum query optimization
Compressed, Archived, Partitioned — three storage strategies
A theorem stating that a distributed system can guarantee at most two of three properties: Consistency, Availability, and Partition Tolerance (Correct answer)
Consistency, Accuracy, Performance — three goals of database design

Correct answer: A theorem stating that a distributed system can guarantee at most two of three properties: Consistency, Availability, and Partition Tolerance

The CAP theorem states that in a distributed system, you can only simultaneously guarantee two of three properties: Consistency, Availability, and Partition Tolerance.

Question 77: What is the key distinction between synchronous and asynchronous data integration?

Synchronous integration is used only for internal systems while asynchronous is reserved for external API calls
Asynchronous integration always produces higher data quality outcomes than synchronous integration
Synchronous integration uses batch processing while asynchronous integration only uses real-time streaming
In synchronous integration the sender waits for a response before proceeding; in asynchronous the sender continues without waiting (Correct answer)

Correct answer: In synchronous integration the sender waits for a response before proceeding; in asynchronous the sender continues without waiting

Synchronous integration requires the calling system to wait for a response before proceeding, while asynchronous integration allows the sender to continue processing after sending a message, receiving any response later.

Question 78: Which technique uses statistical analysis to identify records that deviate significantly from expected patterns, indicating potential quality issues?

Outlier Detection / Anomaly Analysis (Correct answer)
Data Standardization
Data Masking
Record Linkage

Correct answer: Outlier Detection / Anomaly Analysis

Outlier and anomaly detection uses statistical methods to identify data values or records that fall outside expected ranges, signaling potential quality issues.

Question 79: Which financial data concept describes the process of matching trade confirmations between a buyer and seller to verify terms?

Trade confirmation/affirmation (matching) (Correct answer)
Settlement finality
Netting
Novation

Correct answer: Trade confirmation/affirmation (matching)

Trade matching compares buyer and seller confirmations to verify that both parties agree on the trade terms before settlement.

Question 80: What is 'data standardization' in data quality management?

Defining the minimum acceptable quality thresholds for data
Selecting a single DBMS vendor for the organization
Creating documentation standards for data assets
Converting data to a common format, structure, or value set to ensure consistency (Correct answer)

Correct answer: Converting data to a common format, structure, or value set to ensure consistency

Data standardization converts data to consistent formats, structures, and representations (e.g., date formats, address formats, abbreviation expansion) to enable accurate comparison and integration.

Question 81: What is 'data provenance'?

The geographic origin of data storage servers
The process of archiving old data
The documented history of data's origin and transformations over its lifecycle (Correct answer)
A method for encrypting sensitive data

Correct answer: The documented history of data's origin and transformations over its lifecycle

Data provenance documents the origin, custody, and transformations of data throughout its lifecycle to ensure trustworthiness and traceability.

Question 82: What does the integration pattern known as 'scatter-gather' accomplish?

Archiving data by distributing it across geographically dispersed data centers
Randomly distributing data partitions across multiple storage nodes for load balancing
Encrypting data by splitting the payload across multiple secure transmission channels
Broadcasting a request to multiple services simultaneously and then aggregating all responses into one result (Correct answer)

Correct answer: Broadcasting a request to multiple services simultaneously and then aggregating all responses into one result

The scatter-gather pattern fans out a single request to multiple services concurrently (scatter), then waits for and combines all responses into a single aggregated result (gather).

Question 83: Which concept refers to a single, authoritative definition of a business term used consistently across an organization?

Authoritative Data Source
Data Dictionary (Correct answer)
Master Data Record
Golden Record

Correct answer: Data Dictionary

A Data Dictionary provides authoritative definitions of business terms and data elements to ensure consistent understanding across the enterprise.

Question 84: What is 'transparent data encryption' (TDE) in databases?

Encrypting the entire database storage files at rest so data is protected if storage media is stolen (Correct answer)
Encrypting data during transmission over the network
Making encryption keys visible to database administrators
Encrypting only specific sensitive columns in a table

Correct answer: Encrypting the entire database storage files at rest so data is protected if storage media is stolen

TDE encrypts database storage files at rest, protecting data if physical storage media is stolen or accessed without authorization, transparently to applications.

Question 85: Using a weighted average in

Regression
Classification
Above All
Forecasting (Correct answer)

Correct answer: Forecasting

Weighted average is indeed used in forecasting. In forecasting, a weighted average is used to assign different weights or importance to different data points or periods based on their significance or relevance to the forecasted value.

Question 86: In financial data management, what is a 'mark-to-market' valuation?

Adjusting inventory to lower of cost or net realizable value
Spreading asset cost over its useful life
Recording revenue when payment is received
Valuing an asset at its current market price rather than historical cost (Correct answer)

Correct answer: Valuing an asset at its current market price rather than historical cost

Mark-to-market (fair value accounting) records assets and liabilities at their current market value each reporting period.

Question 87: A strategy for earning and spending cash.

Income
Budget (Correct answer)
Expenses
Sales Tax

Correct answer: Budget

A budget is a plan for making and spending money. It helps individuals, households, businesses, and organizations manage their finances effectively by allocating income to various expenses and savings goals.

Question 88: What is the purpose of a Business Glossary in data governance?

Provide standardized business term definitions used across the enterprise (Correct answer)
Monitor database performance
Store raw transactional data
Track ETL job schedules

Correct answer: Provide standardized business term definitions used across the enterprise

A Business Glossary provides a common vocabulary with standardized definitions to ensure consistent communication across business units.

Question 89: Which data quality dimension is most critical when integrating real-time market tick data for algorithmic trading?

Completeness
Conformity
Uniqueness
Timeliness (Correct answer)

Correct answer: Timeliness

In algorithmic trading, stale or delayed tick data can lead to incorrect signals and costly trading errors, making timeliness paramount.

Question 90: Which financial statement reports a company's revenues, expenses, and net income over a specific period?

Balance Sheet
Income Statement (Correct answer)
Statement of Retained Earnings
Statement of Cash Flows

Correct answer: Income Statement

The Income Statement (Profit & Loss Statement) summarizes revenues and expenses over a defined reporting period.

Question 91: A data analyst notices that a company's balance sheet does not balance. Which accounting identity should hold true?

Assets = Revenue - Liabilities
Assets = Liabilities + Equity (Correct answer)
Equity = Assets + Liabilities
Revenue = Expenses + Profit

Correct answer: Assets = Liabilities + Equity

The fundamental accounting equation states that Assets must equal Liabilities plus Shareholders' Equity.

Question 92: In data modeling, what does cardinality describe?

The number of columns in a table
The size of a database in gigabytes
The numerical relationship between instances of two entities (e.g., one-to-many) (Correct answer)
The number of indexes on a table

Correct answer: The numerical relationship between instances of two entities (e.g., one-to-many)

Cardinality describes the quantitative relationship between entity instances, such as one-to-one, one-to-many, or many-to-many relationships.

Question 93: What is the primary goal of a Master Data Management (MDM) program?

Manage database licenses
Automate ETL workflows
Create a single, consistent, authoritative view of key business entities like customers and products (Correct answer)
Optimize database query performance

Correct answer: Create a single, consistent, authoritative view of key business entities like customers and products

MDM creates a single, trusted, authoritative record (golden record) for key business entities to eliminate duplicates and inconsistencies across systems.

Question 94: What is the purpose of a canonical data model in an enterprise integration architecture?

To define a common data format that reduces point-to-point mapping between systems (Correct answer)
To compress data for more efficient storage in the data warehouse
To enforce database security roles and access permissions across systems
To provide audit trails required for regulatory compliance reporting

Correct answer: To define a common data format that reduces point-to-point mapping between systems

A canonical data model defines a shared, standardized format so each system only needs to map to/from the canonical model rather than maintaining a unique mapping to every other system.

Question 95: What does 'data deduplication' specifically address?

Compressing redundant data blocks in storage systems
Identifying and removing or merging duplicate records representing the same real-world entity (Correct answer)
Creating backup copies of critical data
Encrypting duplicate copies of data for security

Correct answer: Identifying and removing or merging duplicate records representing the same real-world entity

Data deduplication identifies records that represent the same entity (e.g., the same customer) and removes or merges them to create a single, accurate record.

Question 96: In data integration, what does 'data latency' refer to?

The number of transformation steps applied to data
The delay between when source data changes and when it becomes available in the target system (Correct answer)
The volume of data processed per hour by the integration pipeline
The encryption strength applied during data transfer

Correct answer: The delay between when source data changes and when it becomes available in the target system

Data latency is the time delay between a change occurring in the source system and that change being reflected and available in the destination system.

Question 97: What is 'connection pooling' in database applications?

Reusing a pool of pre-established database connections to reduce the overhead of creating new connections for each request (Correct answer)
Distributing database load across multiple storage pools
Combining multiple database servers into a cluster
Pooling query results for caching purposes

Correct answer: Reusing a pool of pre-established database connections to reduce the overhead of creating new connections for each request

Connection pooling maintains a pool of reusable database connections, avoiding the costly overhead of establishing a new connection for each application request.

Question 98: A graph that displays data using horizontal or vertical bars.

Bar Graph (Correct answer)
Line Graph
Pie chart
Histogram

Correct answer: Bar Graph

A bar graph is a type of chart that uses horizontal or vertical bars to represent and compare data categories. The length or height of each bar corresponds to the value or quantity being represented. Bar graphs are commonly used to display categorical data and make it easy to visualize comparisons between different categories or groups.

Question 99: Using sequential modeling,

KNN
ANN
RNN (Correct answer)
CNN

Correct answer: RNN

Sequential modeling is indeed done on Recurrent Neural Networks (RNNs). RNNs are a type of neural network architecture specifically designed to work with sequential data, such as time series or text data.

Question 100: What is the key difference between a data policy and a data standard in governance?

Policies apply to databases only; standards apply to all systems
Policies state what must be done; standards specify how to do it (Correct answer)
Policies are technical, standards are business-focused
Standards are mandatory; policies are optional

Correct answer: Policies state what must be done; standards specify how to do it

Data policies state high-level requirements and what must be achieved, while data standards specify the detailed, measurable methods for meeting those policies.

Certified Data Professional (CDP)

The CDP, offered by the Institute for Certification of Computing Professionals (ICCP), validates expertise in the DMBOK framework across core data management disciplines including governance, architecture, integration, and quality. It is recognized as a professional-level credential for data practitioners managing enterprise data assets.

Questions

35 min

Time Limit

70%

To Pass

Checking for saved progress...

Exam Rules

You can skip questions and return to them later
Flag questions for review before submitting
No feedback shown until you submit the entire exam
Unanswered questions count as wrong — answer everything
10 pretest questions are mixed in and don't affect your score
Timer auto-submits when time runs out
Your progress is auto-saved every 30 seconds