AWS DEA-C01 Data Engineer – Associate Practice Exam
Prepare for the AWS Certified Data Engineer – Associate exam (DEA-C01) with scenario-based, source-cited practice questions across all four domains — data ingestion and transformation, data store management, operations and support, and security and governance. Covers Glue, Kinesis, Redshift, S3 lakehouse with Iceberg, Lake Formation, DMS, Step Functions, and more. Includes Learn mode and full timed Exam mode.
Start 24-hour free trial →DEA-C01 exam at a glance
- Vendor
- Amazon Web Services (AWS)
- Exam code
- DEA-C01
- Certification
- AWS Certified Data Engineer – Associate
- Level
- Associate
- Format
- 65 questions (50 scored + 15 unscored); multiple choice and multiple response
- Duration
- 130 minutes
- Scoring
- Scaled 100–1000; passing score 720 (pass/fail, compensatory)
- Cost (USD)
- $300 (50% discount voucher available after any prior AWS certification)
- Delivery
- Pearson VUE test center or online proctored
- Prerequisites
- None required. AWS recommends 2–3 years of data-engineering experience and ~1 year of hands-on AWS
- Validity
- 3 years; recertify by passing the current version before expiry
- Blueprint
- DEA-C01 exam guide (2026); verify the current revision before scheduling
Sources: AWS — Data Engineer – Associate certification · AWS Certification exam guides. Confirm the current exam guide revision with AWS before scheduling.
About the DEA-C01 certification
AWS Certified Data Engineer – Associate (DEA-C01) validates the ability to build and operate data pipelines, lakehouses, and analytics platforms on AWS, and to monitor, troubleshoot, and optimize them for cost and performance. It targets practitioners who ingest, transform, store, and serve data at scale — choosing the right AWS service for each ingestion, storage, processing, and serving pattern rather than hand-rolling infrastructure. The questions are scenario-based: you read a data-pipeline situation and pick the service or design that meets the stated requirements for latency, retention, concurrency, governance, or cost.
The scope centers on the AWS analytics portfolio: AWS Glue for ETL and cataloging, Amazon Kinesis and Amazon Data Firehose and Amazon MSK for streaming ingestion, Amazon Redshift for warehousing, Amazon S3 with Apache Iceberg for the lakehouse, AWS Lake Formation for fine-grained governance, AWS Step Functions for orchestration, AWS DMS for change-data-capture, and Amazon Macie for sensitive-data discovery. A common testing pattern is distinguishing similar services — Kinesis Data Streams vs Firehose vs MSK, or Glue jobs vs crawlers vs the Data Catalog. For the concepts behind these scenarios, see the AWS data-engineering study guides in the Learning Hub.
DEA-C01 exam domains and weights
AWS publishes four scored content domains with the official weightings below (they sum to 100%). Data Ingestion and Transformation is by far the heaviest domain, so plan your study time in proportion — but note the exam uses compensatory scoring, so you only need to pass overall, not each domain.
Build streaming and batch pipelines with AWS Glue, AWS Lambda, Amazon Kinesis Data Streams, Amazon Data Firehose, and Amazon MSK; transform with Apache Spark and SQL; orchestrate pipelines and apply programming concepts.
Design Amazon S3 lakehouse layouts with Apache Iceberg, model warehouses in Amazon Redshift, choose between Amazon RDS, Amazon Aurora, and Amazon DynamoDB, and govern the catalog with AWS Lake Formation. Includes partition-strategy design.
Monitor pipelines with Amazon CloudWatch, orchestrate with AWS Step Functions, validate with AWS Glue Data Quality, remediate failed jobs, and automate and analyze data with AWS services.
Apply IAM authentication and authorization for data, configure row- and column-level security in AWS Lake Formation, encrypt and mask with AWS KMS, prepare audit logs, and detect sensitive data with Amazon Macie.
Source: AWS Certified Data Engineer – Associate (DEA-C01) exam guide. Weights are AWS's published scored-content percentages and sum to 100%.
Who DEA-C01 is for
This associate exam fits practitioners who build and run data platforms on AWS:
- Data engineers building and operating ingestion, transformation, and serving pipelines on AWS.
- Analytics and ETL engineers working with Glue, Redshift, Athena, and the S3 lakehouse.
- Data platform engineers responsible for pipeline reliability, governance, and cost/performance optimization.
- Cloud and software engineers moving into data engineering who already know SQL and Python or Spark.
There are no mandatory prerequisites, but AWS recommends two to three years of data-engineering experience plus about a year of hands-on AWS, including SQL, Python or Spark, and familiarity with the analytics portfolio. Many candidates pair this with the Solutions Architect Associate or move on to the Machine Learning Engineer Associate. For where AWS data-engineering skills lead, see the data engineering career paths in the Career Hub.
What this DEA-C01 practice exam delivers
Learn mode
Get the correct answer, the service-selection reasoning, why each distractor fails, and a link to the relevant AWS documentation right after each question — built for the "choose the right service" scenarios DEA-C01 tests.
Exam mode
Rehearse on a 130-minute timer against 65-question sets that mirror the real DEA-C01 format, so you build the pacing to work through pipeline scenarios under time.
Source-linked explanations
Every answer cites official AWS documentation — Glue, Kinesis, Redshift, Lake Formation, and more — so you study from AWS's own material, not a mystery answer key.
Score by domain
Results break down across the four DEA-C01 domains, so practice tells you exactly which area — ingestion, storage, operations, or security — to sharpen next.
Sample DEA-C01 practice questions
Ten free questions across the four DEA-C01 domains, with full explanations and links to the relevant AWS documentation. The complete bank is available with the 24-hour trial.
A pipeline must ingest streaming clickstream events at ~50,000 events/second, retain them 24 hours, and let two separate consumer applications replay events independently. Which service best fits?
- Amazon SQS standard queue
- Amazon SNS
- Amazon Kinesis Data Firehose only
- Amazon Kinesis Data Streams
Show answer & explanation
Correct: D — Amazon Kinesis Data Streams. KDS retains data (default 24 hours, up to 365 days), supports multiple parallel consumers replaying independently, and scales via shards or on-demand mode.
Why not the others: an SQS standard queue (A) deletes messages after consumption, so there is no cross-consumer replay; SNS (B) is pub/sub without retained, replayable streaming; and Firehose (C) is delivery-only with no independent replay.
Source: AWS — Amazon Kinesis Data Streams →An analytics team needs petabyte-scale columnar storage with sustained concurrent SQL queries and automatic handling of query bursts. Which service fits?
- Amazon Redshift with Concurrency Scaling
- Amazon RDS for PostgreSQL
- Amazon DynamoDB
- Amazon S3 with Athena only
Show answer & explanation
Correct: A — Amazon Redshift with Concurrency Scaling. Redshift is a columnar MPP warehouse, and Concurrency Scaling adds transient capacity to absorb query bursts, making it purpose-built for sustained concurrent BI workloads.
Why not the others: RDS for PostgreSQL (B) is row-oriented OLTP; DynamoDB (C) is key-value NoSQL; and Athena on S3 (D) works but is not purpose-built for sustained, high-concurrency warehouse workloads.
Source: AWS — Redshift Concurrency Scaling →AWS Glue ETL jobs are failing intermittently with "Resource unavailable" errors during peak hours. What is the best first remediation?
- Disable monitoring
- Switch to manual SSH tuning
- Lower job concurrency, or use job bookmarks and retries with exponential backoff
- Move the workload to an EC2 cron job
Show answer & explanation
Correct: C. Reducing concurrent DPU demand and enabling retries handles transient resource pressure, and job bookmarks prevent reprocessing on retry.
Why not the others: disabling monitoring (A) hides the problem; manual SSH tuning (B) is not how the managed service is operated; and moving to an EC2 cron job (D) abandons the managed service unnecessarily.
Source: AWS — AWS Glue Developer Guide → Further reading: PowerKram — Glue ETL reliability →A data engineer needs to land Kinesis stream data into Amazon S3 partitioned by event date, in Parquet format, with minimal code. Which service does this most simply?
- An EMR cluster running 24/7
- A manual Lambda function writing CSV per record
- Amazon Data Firehose with dynamic partitioning and Parquet conversion
- A Glue job triggered every minute
Show answer & explanation
Correct: C — Amazon Data Firehose with dynamic partitioning and Parquet conversion. Firehose natively converts records to Parquet and supports dynamic partitioning by attributes such as event date, with minimal code.
Why not the others: a 24/7 EMR cluster (A) is costly overkill; a hand-rolled Lambda writing CSV (B) is inefficient and unmanaged; and a Glue job every minute (D) is more complex than Firehose for straightforward delivery.
Source: AWS — Firehose dynamic partitioning →A central data lake on Amazon S3 must let multiple teams query specific columns and rows of tables without copying data, enforced across Athena, Redshift Spectrum, and Glue. Which service governs this fine-grained access?
- IAM only, with object-level allow rules
- S3 bucket policies only
- AWS Lake Formation with column- and row-level security
- AWS KMS only
Show answer & explanation
Correct: C — AWS Lake Formation with column- and row-level security. Lake Formation centralizes table, column, and row-level grants that integrated services (Athena, Redshift Spectrum, Glue) enforce.
Why not the others: IAM (A) and S3 bucket policies (B) operate on objects, not table columns and rows; and KMS (D) handles encryption, not access governance.
Source: AWS — AWS Lake Formation → Further reading: PowerKram — data-lake governance →A team needs an OLTP database that auto-scales storage up to 128 TB, offers MySQL/PostgreSQL compatibility, and provides low-latency read replicas. Which service fits?
- Amazon Redshift
- Amazon DynamoDB
- Amazon RDS for SQL Server
- Amazon Aurora
Show answer & explanation
Correct: D — Amazon Aurora. Aurora auto-scales storage up to 128 TB and offers MySQL- and PostgreSQL-compatible engines with low-latency read replicas.
Why not the others: Redshift (A) is a warehouse, not OLTP; DynamoDB (B) is NoSQL key-value; and RDS for SQL Server (C) is a different engine without the MySQL/PostgreSQL compatibility specified.
Source: AWS — Amazon Aurora →An engineer wants to orchestrate a multi-step ETL workflow with retries, branching, and human-approval steps. Which service is purpose-built for this?
- AWS Step Functions
- AWS Lambda alone
- Amazon SQS
- Amazon EventBridge (CloudWatch Events)
Show answer & explanation
Correct: A — AWS Step Functions. Step Functions provides a state machine with retries, choice and parallel states, and human-approval patterns for multi-step workflows.
Why not the others: Lambda alone (B) has no built-in orchestration; SQS (C) is a queue, not a workflow engine; and EventBridge (D) triggers events but does not orchestrate multi-step workflows.
Source: AWS — AWS Step Functions → Further reading: PowerKram — pipeline orchestration →A pipeline needs to capture row-level changes from an on-premises Oracle database into Amazon S3 in near-real-time. Which service is appropriate?
- AWS Snowball Edge
- AWS Database Migration Service (DMS) with CDC and an S3 target
- Amazon SES
- AWS Backup
Show answer & explanation
Correct: B — AWS DMS with CDC and an S3 target. DMS supports change-data-capture from Oracle and writing to S3 (CSV or Parquet) for near-real-time replication.
Why not the others: Snowball Edge (A) is for offline bulk transfer; SES (C) is an email service; and AWS Backup (D) creates backups, not continuous CDC.
Source: AWS — DMS S3 target with CDC → Further reading: PowerKram — change-data-capture on AWS →PII must be discovered automatically across Amazon S3 buckets and reported. Which service does this?
- Amazon Macie
- Amazon GuardDuty
- AWS WAF
- AWS Config
Show answer & explanation
Correct: A — Amazon Macie. Macie uses machine learning and pattern matching to discover sensitive data such as PII in S3 and reports findings.
Why not the others: GuardDuty (B) detects threats; WAF (C) filters web requests; and Config (D) tracks resource configuration — none is a sensitive-data discovery service.
Source: AWS — Amazon Macie → Further reading: PowerKram — AWS Security Specialty (SCS-C02) →A read-heavy Amazon DynamoDB table experiences hot-partition throttling because 80% of reads target a small set of items. Which solution most directly fixes this?
- Lower the table's read capacity units (RCUs)
- Add DynamoDB Accelerator (DAX) for in-memory caching
- Switch to provisioned mode without changing the schema
- Add a global secondary index on the same partition key
Show answer & explanation
Correct: B — Add DynamoDB Accelerator (DAX). DAX caches hot items in memory in front of DynamoDB, dramatically reducing hot-partition read pressure for read-heavy workloads.
Why not the others: lowering RCUs (A) makes throttling worse; changing capacity mode alone (C) does not spread hot keys; and a GSI on the same partition key (D) provides no key distribution.
Source: AWS — Amazon DynamoDB Accelerator (DAX) →Keep going: Learning & Career resources
Data-engineering skills anchor a broad AWS analytics and ML track. Both PowerKram hubs back your next step.
Deep dive: exam format, scoring, study path, and where to go next
Exam format and scoring
DEA-C01 has 65 questions in 130 minutes: 50 are scored and 15 are unscored pretest items that do not affect your result (you cannot tell which are which). Questions are multiple choice (one correct answer) and multiple response (two or more). Scoring is scaled from 100 to 1000, and 720 is the minimum passing score. The exam uses a compensatory model, so you need to pass only the overall exam, not each domain. There is no penalty for guessing. Read the DEA-C01 format guide →
The service-selection testing pattern
DEA-C01 rewards knowing which service fits which pattern. Memorize when to choose Kinesis Data Streams versus Amazon Data Firehose versus Amazon MSK — each maps to a distinct latency, retention, and consumer pattern. Know the difference between Glue jobs, Glue crawlers, and the Glue Data Catalog, because the exam mixes them in distractors. And for Redshift questions, check whether the workload is OLTP or OLAP before answering — Redshift is rarely correct for OLTP scenarios. Read the service-selection primer →
Realistic study path
This exam rewards hands-on depth. A practical plan: work the official exam guide domain by domain, then build real pipelines — a Kinesis-to-S3 Firehose flow with Parquet and dynamic partitioning, a Glue ETL job with bookmarks, a Redshift warehouse with Concurrency Scaling, Lake Formation column- and row-level grants, and a DMS CDC task into S3. Start with the highest-weighted domain (Ingestion and Transformation at 34%) so each study session maximizes score impact, then rehearse in Learn mode and run full 65-question, 130-minute simulations in Exam mode. Read the DEA-C01 study plan →
Where to go next
Data engineering pairs naturally with adjacent AWS credentials. The Machine Learning Engineer – Associate extends your pipeline skills toward productionizing models, the Machine Learning – Specialty (MLS-C01) goes deeper on modeling, and the Solutions Architect – Associate broadens your architecture foundation. See more AWS exams →
Cost, scheduling, and retakes
The exam costs $300 USD and is delivered at a Pearson VUE test center or online with a proctor. If you have earned any prior AWS certification, a 50% discount voucher is available in your AWS Certification account. If you do not pass, AWS requires a 14-day wait before retaking, with no annual cap. The certification is valid for three years; recertify by passing the current version before it expires. Career Hub — data engineering roles →
Frequently asked questions
What format is the DEA-C01 exam?
65 questions in 130 minutes — 50 scored and 15 unscored pretest items. Questions are multiple choice and multiple response. Scoring is scaled from 100 to 1000, and the minimum passing score is 720. The exam uses compensatory scoring, so you need to pass only the overall exam, not each domain.
What does DEA-C01 cover?
Four domains: Data Ingestion and Transformation (34%), Data Store Management (26%), Data Operations and Support (22%), and Data Security and Governance (18%). Topics include AWS Glue, Kinesis, Amazon Data Firehose, MSK, Redshift, the S3 lakehouse with Apache Iceberg, Lake Formation, Step Functions, DMS, and Macie.
Are there prerequisites for DEA-C01?
No mandatory prerequisites. AWS recommends two to three years of data-engineering experience plus about one year of hands-on AWS, including SQL, Python or Spark, and familiarity with the AWS analytics service portfolio.
How much does DEA-C01 cost and how long is it valid?
The exam costs $300 USD (a 50% discount voucher is available after any prior AWS certification). The certification is valid for three years; you recertify by passing the current version of the exam before it expires.
What should I pursue after DEA-C01?
Data engineering pairs well with adjacent AWS credentials: the Machine Learning Engineer – Associate for productionizing models, the Machine Learning – Specialty (MLS-C01) for deeper modeling, or the Solutions Architect – Associate (SAA-C03) for a broader architecture foundation.
Start your free 24-hour DEA-C01 practice trial
Full access to 700+ questions, both study modes, source-linked explanations, and score-by-domain. No credit card required.
Start free trial →