Skip to content

Commit 073c460

Browse files
author
Bob Strahan
committed
> Change Knowledge Base default vector store from OpenSearch Serverless to S3 Vectors for cost optimization
1 parent 912ff8b commit 073c460

File tree

3 files changed

+11
-5
lines changed

3 files changed

+11
-5
lines changed

CHANGELOG.md

Lines changed: 6 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -55,6 +55,12 @@ SPDX-License-Identifier: MIT-0
5555
- removed image limits from multimodal inference steps (classification, extraction, assessment) following Amazon Bedrock API removal of image count restrictions. The system now processes all document pages without artificial truncation, with info logging to track image counts for monitoring purposes.
5656
- Resolves #147
5757

58+
- **Knowledge Base Vector Store Default Changed to S3 Vectors**
59+
- Changed default `KnowledgeBaseVectorStore` from `OPENSEARCH_SERVERLESS` to `S3_VECTORS` for cost-optimized deployments
60+
- S3 Vectors provides 40-60% lower storage costs with sub-second latency suitable for most use cases
61+
- OpenSearch Serverless remains available for applications requiring sub-millisecond query performance
62+
- No action required for existing deployments - only affects new stack deployments
63+
5864
### Fixed
5965

6066
- **Document Schema Builder Enum Support** - Fixed enum value handling in schema builder to properly support enumeration constraints for attribute definitions

docs/knowledge-base.md

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -55,8 +55,8 @@ The Knowledge Base feature supports two vector storage backends to optimize for
5555

5656
### Choosing Your Vector Store
5757

58-
- **OpenSearch Serverless** (Default): Choose for applications requiring ultra-fast retrieval and real-time performance
59-
- **S3 Vectors**: Choose for cost optimization when query latency is acceptable
58+
- **S3 Vectors** (Default): Choose for cost optimization with acceptable sub-second query latency
59+
- **OpenSearch Serverless**: Choose for applications requiring ultra-fast retrieval and real-time performance
6060

6161
## Configuration
6262

@@ -73,7 +73,7 @@ ShouldUseDocumentKnowledgeBase:
7373

7474
KnowledgeBaseVectorStore:
7575
Type: String
76-
Default: "OPENSEARCH_SERVERLESS"
76+
Default: "S3_VECTORS"
7777
AllowedValues:
7878
- "OPENSEARCH_SERVERLESS"
7979
- "S3_VECTORS"

template.yaml

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -230,11 +230,11 @@ Parameters:
230230

231231
KnowledgeBaseVectorStore:
232232
Type: String
233-
Default: "OPENSEARCH_SERVERLESS"
233+
Default: "S3_VECTORS"
234234
AllowedValues:
235235
- "OPENSEARCH_SERVERLESS"
236236
- "S3_VECTORS"
237-
Description: Choose vector store type for Bedrock Knowledge Base. OpenSearch Serverless (default) for sub-millisecond queries, S3 Vectors for cost-effective storage with sub-second latency.
237+
Description: Choose vector store type for Bedrock Knowledge Base. S3 Vectors (default) for cost-effective storage with sub-second latency, OpenSearch Serverless for sub-millisecond queries.
238238

239239
KnowledgeBaseModelId:
240240
Type: String

0 commit comments

Comments
 (0)