CloudGuild · Blog · Cheat sheets · Lessons · Certifications
Navigating AWS Data Lake Architecture Questions for SAA-C03
Learn why selecting services for a data lake can trip up AWS Certified Solutions Architect - Associate candidates and how to choose correctly.
A common stumbling block for candidates is understanding the services that best support a data lake architecture. The options often include a mix of database services and storage solutions, making it tricky to identify the right combination for both structured and unstructured data.
The question
A startup wants to implement a data lake architecture to analyze large datasets from various sources. They want a solution that supports both structured and unstructured data. Which AWS services should they use?
A. Amazon RDS and Amazon Redshift
B. Amazon S3 and AWS Glue
C. AWS Snowball and Amazon DynamoDB
D. Amazon Aurora and Amazon Elasticsearch Service
Think before you scroll
Consider the requirements for a data lake: it must handle both structured and unstructured data efficiently. The right services need to support diverse data formats and offer capabilities for data cataloging and preparation.
The answer
The correct option is B. Amazon S3 and AWS Glue. Amazon S3 serves as an ideal storage solution for a data lake, accommodating both structured and unstructured data. AWS Glue complements this by cataloging and preparing the data for analysis, ensuring you have a flexible and efficient setup.
Why the other options lose
A. Amazon RDS and Amazon Redshift: These services are primarily designed for structured data. RDS is a relational database service, and Redshift is a data warehouse solution. They lack the necessary flexibility for unstructured data storage, making them unsuitable for a data lake architecture.
C. AWS Snowball and Amazon DynamoDB: Snowball is a data transfer service, not a storage solution. While DynamoDB can store structured data, it does not support the unstructured data component required for a data lake. Therefore, this combination fails to meet the requirements.
D. Amazon Aurora and Amazon Elasticsearch Service: Aurora is a relational database service, similar to RDS, and is not designed for unstructured data. While Elasticsearch can handle semi-structured data, it is not a complete solution for a data lake. This combination also does not fulfill the necessary criteria.
The concept behind it
Understanding the distinction between structured and unstructured data is crucial. A data lake architecture requires services that can accommodate both. Amazon S3 is the leading choice for storage, while AWS Glue provides the tools to prepare and analyze that data effectively. This principle applies to any scenario where diverse data formats are involved.
Exam trap to remember
Remember: For a data lake, think storage first with S3, and cataloging with Glue. This combination is key for handling various data types efficiently.