{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreic5nvlleyjdkvsope4o54edull66qvlvh7ygbixir2kdnfaanudym",
    "uri": "at://did:plc:25rdn5elo5izoxrmtis34zuk/app.bsky.feed.post/3mplv5qc2gal2"
  },
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreifigwytgfkxkg6yltamk6n5rrgkkore7xd5np6q6upqw6riz2nsjy"
    },
    "mimeType": "image/webp",
    "size": 289148
  },
  "path": "/bijkler/aws-bedrock-managed-knowledge-bases-should-we-use-them-3119",
  "publishedAt": "2026-07-01T15:53:28.000Z",
  "site": "https://dev.to",
  "tags": [
    "aws",
    "bedrock",
    "knowledgebase",
    "pipelineburst",
    "stacks section"
  ],
  "textContent": "AWS released **Managed Knowledge Bases for Amazon Bedrock** on 17 June 2026. The feature significantly reduces the operational complexity of building Retrieval-Augmented Generation (RAG) solutions by allowing Bedrock to manage the vector storage, indexing, embeddings, and retrieval infrastructure on your behalf.\n\nFor teams looking to deliver an Agent Core proof of concept or their first production RAG workload quickly, this can be a compelling option. However, there are some important trade-offs to understand before committing to the managed approach.\n\nTraditionally, a Bedrock Knowledge Base required a customer-managed vector store such as:\n\n  * OpenSearch Serverless\n  * OpenSearch Managed Clusters\n  * Aurora PostgreSQL with pgvector\n  * Pinecone\n  * DocumentDB\n  * Other supported vector databases With a Managed Knowledge Base, Bedrock handles the underlying vector infrastructure and embedding model selection for you.\n\n\n\nCreating one from the AWS CLI is straightforward:\n\n\n\n    aws bedrock-agent create-knowledge-base \\\n          --name \"my-managed-kb\" \\\n          --role-arn \"arn:aws:iam::${AWS_ACCOUNT_ID}:role/service-role/AmazonBedrockExecutionRoleForKnowledgeBase_ihv1p\" \\\n          --knowledge-base-configuration '{\n            \"type\": \"MANAGED\",\n            \"managedKnowledgeBaseConfiguration\": {\n              \"embeddingModelType\": \"MANAGED\"\n            }\n          }'\n\n\n#  Advantages\n\n##  Lower operational overhead\n\nThere is no need to provision, secure, monitor, patch, or scale a separate vector database.\n\n##  Lower costs\n\nS3 storage is cheaper than database storage. Pay only for each ingestion and retrieval operation. Indexing and searching compute is free. No 24/7 server costs.\n\n##  Faster time-to-value\n\nManaged Knowledge Bases make it possible to stand up a RAG solution in minutes rather than days.\n\n##  Automatic embedding management\n\nBedrock manages embedding selection and indexing, reducing the number of architectural decisions required from development teams.\n\n##  Cost-effective for smaller workloads\n\nThe managed model can be attractive for:\n\n  * Proofs of Concept\n  * Departmental knowledge bases\n  * Agent Core pilots\n  * Workloads with highly variable usage patterns\n\n\n\n##  Good fit for multiple small knowledge bases\n\nOrganizations experimenting with several independent knowledge domains can often benefit from the simplified operational model.\n\n#  Disadvantages\n\n##  Less control\n\nThe primary trade-off is reduced visibility and control over the underlying vector store and retrieval implementation.\n\n##  Potential performance limitations\n\nOrganizations with extremely high query-per-second (QPS) requirements may still prefer dedicated OpenSearch-based solutions where indexing, scaling, and query performance can be tuned explicitly.\n\n##  Fewer advanced search capabilities\n\nIf your use case requires sophisticated filtering, custom ranking strategies, or hybrid lexical/vector search, a self-managed OpenSearch solution may remain the better choice.\n\n##  Infrastructure-as-Code support is not yet mature\n\nAt the time of writing, Managed Knowledge Bases are available through the Bedrock APIs and Console, but CloudFormation and CDK support have not yet fully caught up.\n\n#  CloudFormation and CDK Considerations\n\nOne of the biggest surprises for infrastructure engineers is the current gap between the Bedrock APIs and CloudFormation support.\n\nWhile a Managed Knowledge Base can be created through the Bedrock API and AWS CLI, there is currently no equivalent CloudFormation resource definition that maps cleanly to the managed configuration. This means CDK users cannot yet create Managed Knowledge Bases using standard CDK constructs.\n\nFor teams that require full Infrastructure-as-Code automation, a common workaround is a Lambda-backed Custom Resource that invokes the Bedrock API during deployment.\n\nThis situation is not unusual for newly released AWS services. Historically, CloudFormation and CDK support often follow shortly after the underlying service APIs become available.\n\n#  Recommendation\n\nIf your goal is to deliver a RAG-enabled Agent Core solution quickly and with minimal operational overhead, a Managed Knowledge Base is currently the most attractive starting point.\n\nIf, however, your organization requires:\n\n  * Advanced search capabilities\n  * Full infrastructure control\n  * Sophisticated indexing strategies\n  * Complete CloudFormation/CDK automation\n\n\n\n... then a traditional OpenSearch-based Knowledge Base remains the more flexible option.\n\n#  TL;DR\n\nUse Managed Knowledge Bases when speed, simplicity, and lower operational overhead are more important than infrastructure control.\n\n#  Postscript: Cloud Formation and CDK Engineering Insights\n\nAs with most new AWS Products, the Cloud Formation types and CDK support comes at some later date; forget including this in your codebase unless you want to spend the time writing a Lambda backed Custom Resource.\n\nYou might find it interesting that the original OpenSearch Serverless Vector Collection also requires a Custom Resource in CDK. The reason is the cluster is created using a Cloud Formation type but there are no default indexes defined. An \"on-the-fly\"\nLambda is created which initialises the cluster with an index that the Knowledge Base can use.\n\nIf you want to take a deep dive into this approach try Dirk Michel's pipelineburst example from his Bedrock examples repository in his CDK stacks section (written about 2 years ago).",
  "title": "AWS Bedrock Managed Knowledge Bases: Should We Use Them?"
}