AIP-C01 · Question #47
A company provides a service that helps users from around the world discover new restaurants. The service has 50 million monthly active users. The company wants to implement a semantic search…
The correct answer is B. Migrate the restaurant data to Amazon OpenSearch Service. Use a foundation model (FM) in. Amazon OpenSearch Service with a foundation model for vector embeddings and semantic search is the best fit. OpenSearch natively supports k-NN vector search (ANN indexing with FAISS/NMSLIB), scales horizontally to handle 20M documents and high query throughput, and supports…
Question
A company provides a service that helps users from around the world discover new restaurants. The service has 50 million monthly active users. The company wants to implement a semantic search solution across a database that contains 20 million restaurants and 200 million reviews. The company currently stores the data in PostgreSQL. The solution must support complex natural language queries and return results for at least 95% of queries within 500 ms. The solution must maintain data freshness for restaurant details that update hourly. The solution must also scale cost-effectively during peak usage periods. Which solution will meet these requirements with the LEAST development effort?
Options
- AMigrate the restaurant data to Amazon OpenSearch Service. Implement keyword-based search
- BMigrate the restaurant data to Amazon OpenSearch Service. Use a foundation model (FM) in
- CKeep the restaurant data in PostgreSQL and implement a pgvector extension. Use a foundation
- DMigrate restaurant data to an Amazon Bedrock knowledge base by using a custom ingestion
How the community answered
(28 responses)- A25% (7)
- B61% (17)
- C7% (2)
- D7% (2)
Explanation
Amazon OpenSearch Service with a foundation model for vector embeddings and semantic search is the best fit. OpenSearch natively supports k-NN vector search (ANN indexing with FAISS/NMSLIB), scales horizontally to handle 20M documents and high query throughput, and supports near-real-time index updates that meet the hourly freshness requirement. Using an FM to generate embeddings enables true semantic understanding of natural language queries. The other options fall short: (A) OpenSearch with only keyword-based search does not satisfy the semantic/natural language query requirement; (C) pgvector on PostgreSQL cannot cost-effectively scale to handle 50M monthly active users with sub-500ms p95 at this data volume without significant sharding complexity; (D) a Bedrock knowledge base with custom ingestion introduces more development effort and is not optimized for the scale and freshness requirements described.
Topics
Community Discussion
No community discussion yet for this question.