AWS Certified Data Engineer - Associate (DEA-C01) Practice Exam

What the AWS Certified Data Engineer - Associate (DEA-C01) exam covers

• Data Ingestion and Transformation280 questions
• Data Store Management215 questions
• Data Operations and Support181 questions
• Data Security and Governance149 questions

Free DEA-C01 practice test questions

A sample of 10 questions with answers and explanations. A free account includes a fixed set of 20 practice questions for this exam, retryable anytime. Paid access unlocks the full 825-question bank.

1. Question 1Data Ingestion and Transformation

   An analyst wants near-real-time indexing of application logs into a search cluster for operational dashboards, tolerating a delivery latency of about one minute. Which Firehose destination configuration is designed for this?

   • AOpenSearch Service as the destinationCorrect
   • BAmazon S3 with Parquet conversion enabled
   • CAmazon Redshift as the destination
   • DAn HTTP endpoint to a batch reporting tool
   ✓ Correct answer: A

   Firehose has a native OpenSearch Service destination that indexes incoming records into a specified domain and index, which is well suited to operational log analytics dashboards that need data available within roughly a minute rather than requiring true sub-second streaming.

   Why the other options are wrong

   • BS3 with Parquet conversion is intended for cost-efficient analytical storage and batch querying, not live search dashboards.
   • CRedshift is optimized for structured analytical queries via SQL, not for the search-style indexing needed by log dashboards.
   • DAn HTTP endpoint to a batch reporting tool implies periodic batch consumption rather than near-real-time search indexing.
2. Question 2Data Ingestion and Transformation

   A Glue ETL job must extract data from an Amazon RDS database running inside a private VPC subnet. What should the engineer configure so the Glue job can reach the database securely?

   • AA larger DPU allocation assigned to run the ETL job
   • BA public S3 bucket policy granting Glue read access
   • CA Glue connection referencing the RDS VPC and subnetCorrect
   • DA crawler classifier configured specifically for JDBC
   ✓ Correct answer: C

   Glue connections store networking details, including VPC, subnet, and security group information along with credentials, that Glue uses to launch elastic network interfaces inside the VPC so a job can privately reach resources such as an RDS instance. An S3 bucket policy is unrelated to reaching a private database, JDBC classifiers only affect crawler schema parsing, and DPU count affects compute capacity, not network reachability.

   Why the other options are wrong

   • AAdding more DPUs increases compute capacity for the job but does not grant network access into the VPC.
   • BAn S3 bucket policy controls access to S3 objects, it has no effect on reaching a private RDS instance.
   • DA JDBC classifier helps a crawler parse a JDBC source's schema, it does not establish network connectivity.
3. Question 3Data Ingestion and Transformation

   A compliance team requires proof that files copied by AWS DataSync from an on-premises server to Amazon S3 were transferred without corruption. Which built-in DataSync capability provides this?

   • AAutomatic virus scanning that quarantines infected files before transfer
   • BAutomatic encryption key rotation performed during every file transfer
   • CAutomatic integrity verification that compares checksums after transferCorrect
   • DAutomatic schema validation that compares source and target table structures
   ✓ Correct answer: C

   DataSync performs built-in verification, comparing metadata or content between the source and destination to confirm that files were transferred completely and without corruption, and it reports any inconsistencies so they can be investigated.

   Why the other options are wrong

   • ADataSync does not provide virus scanning of transferred files; that capability is not part of its data movement verification feature.
   • BDataSync encrypts data in transit using TLS, but automatic encryption key rotation is not the mechanism it uses to prove data integrity.
   • DSchema validation between database tables is a database migration concept associated with tools like DMS, not a DataSync file transfer capability.
4. Question 4Data Ingestion and Transformation

   A research lab has 200 TB of data on-premises and only a slow, shared internet connection that would take months to upload. Which AWS service lets the lab physically ship the data to AWS instead?

   • AAWS DataSync, an online transfer service using an on-prem agent
   • BAWS Snowball Edge, a rugged device shipped to and from the siteCorrect
   • CAWS Transfer Family, a managed SFTP and FTPS ingestion service
   • DAmazon S3 Transfer Acceleration, a faster upload path over the internet
   ✓ Correct answer: B

   Snowball Edge devices are shipped to the customer's location, loaded with data locally, and shipped back to AWS where the data is ingested into S3, which avoids relying on a slow or limited internet connection for very large one-time transfers.

   Why the other options are wrong

   • AAWS DataSync moves data online over the network through an on-premises agent, which is exactly the constraint the slow connection makes impractical here.
   • CAWS Transfer Family accepts file uploads over SFTP or FTPS across the network, which still depends on the same slow internet connection.
   • DAmazon S3 Transfer Acceleration speeds up network uploads using edge locations, but it still requires transferring the data over the internet.
5. Question 5Data Store Management

   During an upload of a very large object, a network interruption causes one part of a multipart upload to fail partway through. What is the effect on the overall upload if the application retries only the failed part?

   • AAll previously completed parts expire and must all be re-sent
   • BThe entire object must be re-uploaded again starting from the first part
   • COnly the failed part needs to be re-uploaded to complete the objectCorrect
   • DThe upload is automatically aborted with no way to resume progress
   ✓ Correct answer: C

   Each part of a multipart upload is uploaded and acknowledged independently, so a network failure on one part only requires that specific part to be resent before calling CompleteMultipartUpload. This resilience, along with the ability to upload parts in parallel, is a major reason AWS recommends multipart upload for large objects. Incomplete uploads do persist until completed, aborted, or removed by a lifecycle rule, but a single part failure does not by itself abort the whole upload.

   Why the other options are wrong

   • ASuccessfully uploaded parts remain available to complete the multipart upload; they do not expire simply because another part failed.
   • BMultipart upload is designed specifically so a failure in one part does not require re-uploading parts that already succeeded.
   • DAn in-progress multipart upload is not automatically aborted by a single part failure; the application can retry that part and continue.
6. Question 6Data Store Management

   A company wants near real-time analytics on operational data from Amazon Aurora without building and maintaining a custom extract-transform-load pipeline into Redshift. Which capability is designed for this purpose?

   • AA zero-ETL integration replicating Aurora into Redshift liveCorrect
   • BA manual UNLOAD and COPY cycle scheduled to run every few minutes
   • CA federated query pointed at the Aurora cluster for every report run
   • DRedshift Spectrum querying Aurora's exported data files staged in S3
   ✓ Correct answer: A

   Zero-ETL integrations set up managed, continuous replication from a supported source, such as Amazon Aurora, directly into Redshift tables, making recently written operational data available for analytics within seconds to minutes, without the customer building or operating extract, transform, and load jobs.

   Why the other options are wrong

   • BA manual UNLOAD and COPY cycle is exactly the kind of custom pipeline zero-ETL integrations are designed to eliminate.
   • CFederated query runs live cross-database queries on demand rather than continuously replicating data for ongoing near real-time analytics.
   • DAurora does not export to S3 as its analytics mechanism here; zero-ETL integrations replicate directly into Redshift tables instead.
7. Question 7Data Operations and Support

   A Redshift provisioned cluster occasionally bursts to extra concurrency scaling clusters during peak reporting hours. Regarding cost, which statement about concurrency scaling billing is accurate?

   • AConcurrency scaling usage is billed only when the main cluster is paused
   • BConcurrency scaling clusters are billed at a flat monthly rate regardless of use
   • CConcurrency scaling clusters are always free of charge for any provisioned cluster
   • DFree credits accrue daily from cluster usage before per-second charges startCorrect
   ✓ Correct answer: D

   Rather than being free indefinitely or billed as a flat fee, concurrency scaling clusters draw down a pool of credits that accrue based on how much the main cluster runs each day; once those credits are exhausted, additional concurrency scaling usage is billed per second, making this a usage-based cost customers should monitor.

   Why the other options are wrong

   • AConcurrency scaling activates to handle bursts on an active cluster; a paused cluster is not serving queries at all.
   • BBilling is usage-based, accruing per-second beyond the free credit allowance, not a flat monthly rate.
   • CIt is not unconditionally free; free usage is limited to accrued credits, beyond which per-second charges apply.
8. Question 8Data Operations and Support

   A team has several Glue jobs and crawlers with dependencies between them, such as a crawler that must finish before a job starts, and that job's completion triggering two downstream jobs. Which Glue feature groups these related jobs, crawlers, and triggers into a single orchestrated unit that can be monitored as a whole?

   • AA Glue Data Catalog database, which organizes table metadata definitions
   • BA Glue job bookmark, which tracks previously processed source data
   • CA Glue development endpoint, which hosts an interactive coding environment
   • DA Glue workflow, which groups related jobs, crawlers, and triggersCorrect
   ✓ Correct answer: D

   Workflows let you visualize and manage the dependencies between multiple jobs and crawlers, using triggers to define the sequencing, such as starting a job only after a crawler and then fanning out to downstream jobs. The whole workflow can be monitored and run as a single entity.

   Why the other options are wrong

   • AA Data Catalog database stores table and schema metadata and has no role in sequencing job and crawler execution.
   • BA job bookmark tracks processing state for a single job between runs; it does not group multiple jobs and crawlers together.
   • CA development endpoint provides an interactive environment for writing and testing scripts, not an orchestration grouping.
9. Question 9Data Operations and Support

   A nightly Glue crawler re-scans an S3 prefix whose schema and partitions rarely change, and the crawler run cost has become a noticeable line item. What is the MOST cost-effective fix?

   • APoint the crawler's logs to a lower-cost CloudWatch log group
   • BRun the crawler on demand or on a much less frequent scheduleCorrect
   • CHave the crawler classify the data using a custom Grok pattern
   • DConfigure the crawler to use a larger number of concurrent workers
   ✓ Correct answer: B

   Glue crawlers are billed for the compute time they run. If a dataset's schema and partition structure are stable, running the crawler nightly is wasted cost; triggering it on demand or on a much longer interval directly reduces the number of billed crawler runs.

   Why the other options are wrong

   • ALog storage location has negligible effect on the crawler run cost itself.
   • CA custom classifier changes how schema is inferred, not run frequency or cost.
   • DMore concurrency speeds up a single run but does not reduce how often the crawler runs.
10. Question 10Data Security and Governance

    A small team has one S3 bucket used exclusively by a single application, with no need for column-level restrictions or sharing with other teams or accounts. What is the more proportionate access control approach?

    • AA cross-account AWS RAM share even though only one account is involved
    • BA Redshift datashare configured to expose the bucket's objects as tables
    • CA full Lake Formation deployment with LF-tags and data filters applied
    • DA plain IAM policy plus bucket policy scoped to that application's roleCorrect
    ✓ Correct answer: D

    Lake Formation earns its complexity when you need column or row level restrictions, centralized tag-based governance at scale, or controlled sharing across teams and accounts. A single bucket used by one application without those needs is well served by a straightforward IAM policy and bucket policy, avoiding the operational overhead of registering locations and managing catalog-level grants for no added benefit.

    Why the other options are wrong

    • AA cross-account share solves a problem this scenario does not have, since only one account and one application are involved.
    • BA Redshift datashare governs sharing between Redshift clusters, not general application access to a single S3 bucket.
    • CIntroducing LF-tags and data filters adds governance overhead that this simple, single-application scenario does not need.

Who this AWS Certified Data Engineer - Associate (DEA-C01) practice exam is for

This practice set is for anyone preparing for the AWS Certified Data Engineer - Associate (DEA-C01) exam at the intermediate level - from first-time candidates building a foundation to experienced AWS practitioners doing a final review before test day. If you learn best by working through realistic questions and reading why each answer is right or wrong, it is built for you.

How to use this AWS Certified Data Engineer - Associate (DEA-C01) practice exam

1. Start with the free sample questions above to gauge your current baseline.
2. Read the full explanation on every question, including why each wrong option is wrong.
3. Track your weak domains and focus your study where you are losing the most marks.
4. Once you are scoring consistently well, take a timed, full-length mock exam.
5. Use your readiness score to decide when you are ready to book the real AWS Certified Data Engineer - Associate (DEA-C01) exam.

Related AWS resources

• AWS Certified Data Engineer - Associate (DEA-C01) study guideKey concepts
• AWS practice examsAll AWS
• Certification pathWhere this fits
• Certification exam guides & tipsBlog
• Plans & pricingFree & paid
• How these questions are written and reviewedMethodology
• Report a problem with a questionCorrections
• AWS Certified Generative AI Developer - Professional (AIP-C01) practice examRelated
• AWS CLF-C02 practice examRelated
• AWS DOP-C02 practice examRelated

AWS Certified Data Engineer - Associate (DEA-C01) practice exam FAQ