为 Amazon Bedrock 知识库选择向量存储
原标题：Selecting a vector store for Amazon Bedrock Knowledge Bases
AI 摘要
AWS Machine Learning Blog 发布文章，指导用户为 Amazon Bedrock Knowledge Bases 的客户自管配置选择向量存储后端。文章对比了 Amazon OpenSearch Service、Amazon Aurora PostgreSQL with pgvector 和 Amazon S3 Vectors 三种后端，并针对产品目录搜索等 RAG 用例给出选型建议。文中指出 S3 Vectors 可将向量存储成本降低最多 90%，OpenSearch Serverless 适合低延迟混合搜索场景。
正文节选
Selecting a vector store for Amazon Bedrock Knowledge Bases When building a Retrieval Augmented Generation (RAG) solution with Amazon Bedrock Knowledge Bases, selecting the right vector store impacts performance and cost. Amazon Bedrock Knowledge Bases offers a fully managed option and a customer-managed option where you choose your own vector store. This post focuses on the customer-managed path, comparing the three supported backends: Amazon OpenSearch Service, Amazon Aurora PostgreSQL with pg