AWS announces the general availability of vector search for Amazon MemoryDB

AWS announced the general availability of vector search for Amazon MemoryDB, an in-memory database with multi-AZ durability. This feature enables storing, indexing, retrieving, and searching vectors with high performance and recall rates.

Specifically, the article covers:

• Vector search for MemoryDB supports storing millions of vectors with single-digit millisecond query and update latencies at high throughput and >99% recall.
• You can generate vector embeddings using AI/ML services like Amazon Bedrock and Amazon SageMaker, and store them within MemoryDB.
• Vector search for MemoryDB enables developing real-time machine learning (ML) and generative AI applications that require high throughput, high recall rates, and low latency.
• Use cases include detecting anomalies like fraudulent transactions during high transaction volumes with minimal false positives.
• Vector search for MemoryDB is available at no additional cost in all AWS Regions that offer MemoryDB.