DynamoDB holds 99% recall at trillion-vector scale

Amazon DynamoDB added native vector search with single-digit-millisecond latency at 99%+ recall, designed to scale to trillions of vectors. AWS says the new index keeps embeddings beside operational data, removing the separate vector store and synchronization pipeline for agent memory, RAG, and recommendations. Teams already anchored to DynamoDB should benchmark retrieval quality and total cost before renewing a dedicated vector database—the surrounding-service leverage behind Amazon’s $220 billion AI capacity bet just widened.