This is a summary. For a deeper dive, read the LinkedIn article.
I wrote the original note right after passing the AWS Certified Database - Specialty exam. It took about two months of preparation, and the main reason I shared the plan was that the exam was harder than I expected.
The audience for this exam was not someone brand new to databases. The useful baseline was hands-on experience with relational and non-relational databases, plus enough AWS experience to compare services and choose a database design for a real workload.
My preparation approach was simple:
- use one structured course to build the base map
- do hands-on labs where possible
- take practice exams early enough to expose weak areas
- use AWS documentation and cheat sheets to repair the gaps
- revisit services that often appear together in scenario questions
The practice exams were useful, but the real exam felt trickier. The challenge was not only knowing service names. Many questions were about reading the business constraint carefully and mapping it to the right database behavior.
The services and themes I paid the most attention to were:
- RDS, Aurora, DynamoDB, ElastiCache, DMS, and SCT
- high availability and disaster recovery with low RPO and RTO
- migration from on-prem databases to AWS databases
- parameter groups and which changes require restarts
- encryption in transit and at rest
- security groups, IAM roles, and networking configuration
- backup, restore, caching, and engine-specific behavior
- less common services such as Neptune, DocumentDB, and Redshift
The best use of a practice test is not to memorize answers. It is to find where your mental model is incomplete. After each failed question, I would go back to the docs and make sure I understood why the wrong options were wrong.
This note is from the 2022 exam context, so I would verify the current AWS exam guide before using it as a study plan today. The habit still holds: do not study only for recognition. Study until you can explain the tradeoff behind each database choice.