AIP-C01 · Question #81
A financial services company is creating a Retrieval Augmented Generation (RAG) application that uses Amazon Bedrock to generate summaries of market activities. The application relies on a vector…
The correct answer is B. Launch an Amazon MemoryDB cluster and configure the index by using the Hierarchical. Amazon MemoryDB is an in-memory database that delivers microsecond read latency, making it ideal for high-performance vector similarity search. The HNSW (Hierarchical Navigable Small World) algorithm builds a multi-layer graph that provides very high recall accuracy and fast…
Question
A financial services company is creating a Retrieval Augmented Generation (RAG) application that uses Amazon Bedrock to generate summaries of market activities. The application relies on a vector database that stores a small proprietary dataset with a low index count. The application must perform similarity searches. The Amazon Bedrock model's responses must maximize accuracy and maintain high performance. The company needs to configure the vector database and integrate it with the application. Which solution will meet these requirements?
Options
- ALaunch an Amazon MemoryDB cluster and configure the index by using the Flat algorithm.
- BLaunch an Amazon MemoryDB cluster and configure the index by using the Hierarchical
- CLaunch an Amazon Aurora PostgreSQL cluster and configure the index by using the Inverted File
- DLaunch an Amazon DocumentDB cluster that has an IVFFlat index and a high probe value.
How the community answered
(40 responses)- A8% (3)
- B65% (26)
- C23% (9)
- D5% (2)
Explanation
Amazon MemoryDB is an in-memory database that delivers microsecond read latency, making it ideal for high-performance vector similarity search. The HNSW (Hierarchical Navigable Small World) algorithm builds a multi-layer graph that provides very high recall accuracy and fast approximate nearest-neighbor queries, even on small datasets. For a small proprietary dataset where both accuracy and performance are priorities, HNSW on MemoryDB is the optimal choice. Option A uses the Flat algorithm, which is exact but does not scale well in performance. Options C and D use IVFFlat or IVF-based indexes on relational or document databases; IVF indexes require a minimum number of vectors to build effective clusters and are better suited for large datasets, not small low-index-count collections.
Topics
Community Discussion
No community discussion yet for this question.