Skip to content

DEA-C01 - 8-Week Practice Plan

This plan assumes 2-3 hours per day of focused study, 5-6 days per week. Adjust pacing based on your background. If you held DAS-C01 or DBS-C01, compress to 4-5 weeks by skipping topics you're already strong on.

Week 1 - Foundations and Domain 1A (Ingestion)

Goals

  • Solid mental model of the four exam domains
  • Comfortable with all ingestion services

Reading

Hands-on

  • Build: Kinesis Data Firehose → S3 (Parquet) → Glue Crawler → Athena
  • Build: Producer Python script → KDS → Lambda consumer logging to CloudWatch
  • Read AWS docs: Kinesis Data Streams, Firehose, MSK, DMS

Self-check

  • Can I list when to pick KDS vs Firehose vs MSK?
  • Can I describe full-load + CDC with DMS end-to-end?
  • Can I configure Firehose to convert JSON to Parquet?

Week 2 - Domain 1B (Transformation and Orchestration)

Reading

Hands-on

  • Build: Glue ETL job that joins three S3 tables, writes Iceberg, uses job bookmarks
  • Build: Step Functions workflow orchestrating two Glue jobs with retries and a Catch state
  • Compare: same workload in Glue Studio vs Glue notebook vs Glue script

Self-check

  • When do I use DynamicFrame vs DataFrame?
  • When do I pick EMR Serverless vs EMR on EC2 vs Glue?
  • When do I use Step Functions Standard vs Express?
  • When do I use MWAA over Step Functions?

Week 3 - Domain 2A (S3 + Lakehouse)

Reading

  • notes/03-data-stores.md - sections on S3, Iceberg, Hudi, Delta
  • S3 storage classes and lifecycle docs
  • Athena Iceberg docs

Hands-on

  • Configure S3 lifecycle policy for hot/warm/cold tiering
  • Build: convert a JSON dataset to Parquet with Athena CTAS, partition by date
  • Build: an Iceberg table on S3 with Athena, do a time-travel query
  • Use Athena partition projection on a high-cardinality date partition

Self-check

  • What's the right storage class for daily-access logs older than 30 days?
  • Why is Parquet faster and cheaper for analytics?
  • What does Iceberg give me that plain Parquet doesn't?

Week 4 - Domain 2B (Redshift + Operational Stores)

Reading

  • notes/03-data-stores.md - sections on Redshift, RDS, DynamoDB
  • Redshift Database Developer Guide (chapters on distribution and sort keys)
  • DynamoDB best practices guide

Hands-on

  • Build: Redshift cluster, COPY 10 GB Parquet from S3, run analytical queries
  • Try Redshift Serverless and compare cost characteristics
  • Build: DynamoDB table with composite key, GSI, DynamoDB Streams → Lambda
  • Try Aurora Zero-ETL to Redshift (if access)

Self-check

  • Can I describe Redshift distribution styles (EVEN, KEY, ALL) with examples?
  • Can I tell when to use DynamoDB on-demand vs provisioned?
  • Can I list every way to load Redshift (COPY, streaming, Auto Copy, Zero-ETL)?

Week 5 - Domain 3 (Operations and Support)

Reading

Hands-on

  • Build: CloudWatch Alarm on Glue job failure → SNS email
  • Build: Glue Data Quality rule that fails the job if completeness < 95%
  • Build: Athena workgroup with per-query data scan limit
  • Use CloudWatch Logs Insights to debug a Lambda or Glue job error

Self-check

  • Can I list 5 ways to reduce Athena cost?
  • Can I detect a Kinesis consumer lagging behind?
  • Can I write a basic Logs Insights query?

Week 6 - Domain 4 (Security and Governance)

Reading

Hands-on

  • Build: S3 bucket with SSE-KMS using a customer-managed key
  • Build: Lake Formation table with column-level grant excluding ssn
  • Build: Lake Formation row-level filter so analyst sees only their region
  • Run Macie scheduled job on a sample bucket

Self-check

  • Can I describe the IAM + Lake Formation layered permission model?
  • Can I set up cross-account lake sharing without copying data?
  • Can I enforce TLS-only access on an S3 bucket?

Week 7 - Architecture Patterns and Integration

Reading

Hands-on

  • Build: end-to-end CDC pipeline (RDS Postgres → DMS → S3 → Glue → Iceberg → Athena)
  • Build: Streaming pipeline (KDS → Flink → DynamoDB + Firehose → S3)

Self-check

  • Can I sketch each of the 7 architecture patterns from memory?
  • Can I name 3 cost optimizations for each major service?

Week 8 - Mock Exams and Weak-Area Drilldown

Practice exams

Weak-area drilldown

  • Identify 3-5 weakest domains/topics
  • Re-read those notes
  • Build a small lab targeting each weak area
  • Re-test on questions specific to those topics

Final review

  • strategy.md - exam day approach
  • scenarios.md - second pass of all scenarios
  • AWS Whitepapers: Data Analytics Lens (Well-Architected), Big Data Analytics Options on AWS

Schedule the exam

  • Book Pearson VUE testing center or online proctoring
  • Confirm 130-minute slot
  • Plan exam-day logistics (ID, quiet space if online)

Daily routine (suggested)

Time Activity
30 min Read notes / fact-sheet section
60 min Hands-on build
30 min Practice questions on the day's topic
15 min Review the next day's plan

Stop signals (when you're ready)

You're ready to schedule the exam when all are true:

  • You consistently score 80%+ on AWS official sample questions twice in a row
  • You can sketch each of the 7 architecture patterns from memory
  • You can describe domain weights (34/26/22/18) and your strongest/weakest of the four
  • You can name the right service for each "exam trigger" line in the fact-sheet without looking
  • You have hands-on built at least 5 of the 8 weekly labs