AIF-C01 Question 114
Select 3A company is building a recommendation engine that requires storing and querying high-dimensional vector embeddings for similarity searches. Which of the following AWS services can be used to store embeddings within a vector database for this use case?
- A
Amazon OpenSearch Service
- B
Amazon Aurora
- C
Amazon Neptune
- D
Amazon RDS for PostgreSQL
- E
Amazon DynamoDB
Show answer and explanation
Correct answers: A, C, D
Explanation
To store embeddings for similarity searches in vector databases, services like Amazon OpenSearch Service, Amazon Neptune, and Amazon RDS for PostgreSQL are suitable. OpenSearch supports vector-based operations natively, Neptune is ideal for graph-based embeddings, and PostgreSQL can use extensions like pgvector. While Amazon Aurora and DynamoDB are powerful databases, they are not designed specifically for vector embeddings or similarity searches.
- A. Correct.
Amazon OpenSearch Service is a suitable option for storing vector embeddings as it supports nearest neighbor search for vector-based queries, which is essential for similarity searches.
- B. Incorrect.
Amazon Aurora is a relational database and does not natively support vector operations or similarity searches required for handling embeddings.
- C. Correct.
Amazon Neptune is a graph database that can store embeddings and perform similarity searches, especially when embeddings are part of graph-based datasets.
- D. Correct.
Amazon RDS for PostgreSQL is a relational database but supports extensions like pgvector, which can enable the storage and querying of vector embeddings.
- E. Incorrect.
Amazon DynamoDB is a NoSQL database and is not designed specifically for storing or querying vector embeddings or performing similarity searches.