generative-ai-cdk-constructs
Amazon Aurora Vector Store Construct Library
All classes are under active development and subject to non-backward compatible changes or removal in any future version. These are not subject to the Semantic Versioning model. This means that while you may use them, you may need to update your source code when upgrading to a newer version of this package.
| Language | Package |
|---|---|
 TypeScript |
@cdklabs/generative-ai-cdk-constructs |
 Python |
cdklabs.generative_ai_cdk_constructs |
This construct library provides a class that defines a AmazonAuroraVectorStore construct for an Amazon Aurora to be used for a vector store for a Knowledge Base. Additionally, you can utilize fromExistingAuroraVectorStore() method to use your existing Aurora database as a vector DB. AmazonAuroraVectorStore is an L3 resource that creates a VPC with 3 subnets (public, private with NAT Gateway, private without NAT Gateway) and Amazon Aurora Serverless V2 Cluster. The cluster has 1 writer/reader instance with latest supported PostgreSQL version (currently it is 15.5) and having the following cofiguration: min capacity 0.5, max capacity 4. Lambda custom resource executes required pgvector and Amazon Bedrock Knowledge Base SQL queries (see more here) against Aurora cluster during deployment. The secret containing databases credentials is being deployed and securely stored in AWS Secrets Manager. You must specify the same embeddings model that you are going to use in KnowledgeBase construct. Due to the nature of provisioning RDS cluster it takes a long time (over 20-25 minutes) to both deploying and destroying construct so please take this in consideration.
Table of contents
API
See the API documentation.
Amazon Aurora Vector Store
TypeScript
import { amazonaurora, foundation_models } from '@cdklabs/generative-ai-cdk-constructs';
new amazonaurora.AmazonAuroraVectorStore(stack, 'AuroraVectorStore', {
embeddingsModel: foundation_models.BedrockFoundationModel.COHERE_EMBED_ENGLISH_V3,
});
Python
from cdklabs.generative_ai_cdk_constructs import (
amazonaurora,
foundation_models
)
aurora = amazonaurora.AmazonAuroraVectorStore(self, 'AuroraVectorStore',
embeddings_model=foundation_models.BedrockFoundationModel.COHERE_EMBED_ENGLISH_V3,
)
fromExistingAuroraVectorStore()
You can import your existing Aurora DB to be used as a vector DB for a knowledge base. Note - you need to provide clusterIdentifier, databaseName, vpc, secret and auroraSecurityGroupName used in deployment of your existing RDS Amazon Aurora DB, as well as embeddingsModel that you want to be used by a Knowledge Base for chunking. Additionally, your stack’s env needs to contain region and account variables.
TypeScript
import { amazonaurora, foundation_models, bedrock } from '@cdklabs/generative-ai-cdk-constructs';
import * as cdk from 'aws-cdk-lib';
const auroraDb = amazonaurora.AmazonAuroraVectorStore.fromExistingAuroraVectorStore(stack, 'ExistingAuroraVectorStore', {
clusterIdentifier: 'aurora-serverless-vector-cluster',
databaseName: 'bedrock_vector_db',
schemaName: 'bedrock_integration',
tableName: 'bedrock_kb',
vectorField: 'embedding',
textField: 'chunks',
metadataField: 'metadata',
primaryKeyField: 'id',
embeddingsModel: foundation_models.BedrockFoundationModel.COHERE_EMBED_ENGLISH_V3,
vpc: cdk.aws_ec2.Vpc.fromLookup(stack, 'VPC', {
vpcId: 'vpc-0c1a234567ee8bc90',
}),
auroraSecurityGroupId: 'sg-012ef345678c98a76',
secret: cdk.aws_rds.DatabaseSecret.fromSecretCompleteArn(
stack,
'Secret',
cdk.Stack.of(stack).formatArn({
service: 'secretsmanager',
resource: 'secret',
resourceName: 'rds-db-credentials/cluster-1234567890',
region: cdk.Stack.of(stack).region,
account: cdk.Stack.of(stack).account,
arnFormat: cdk.ArnFormat.COLON_RESOURCE_NAME,
}),
),
});
const kb = new bedrock.KnowledgeBase(this, "KnowledgeBase", {
embeddingsModel: foundation_models.BedrockFoundationModel.COHERE_EMBED_ENGLISH_V3,
vectorStore: auroraDb,
instruction:
"Use this knowledge base to answer questions about books. " +
"It contains the full text of novels.",
});
const docBucket = new cdk.aws_s3.Bucket(this, "DocBucket");
new bedrock.S3DataSource(this, "DataSource", {
bucket: docBucket,
knowledgeBase: kb,
dataSourceName: "books",
chunkingStrategy: bedrock.ChunkingStrategy.fixedSize({
maxTokens: 500,
overlapPercentage: 20,
}),
});
Python
from aws_cdk import (
aws_s3 as s3,
aws_rds as rds,
aws_ec2 as ec2,
Stack,
ArnFormat
)
from cdklabs.generative_ai_cdk_constructs import (
bedrock,
amazonaurora,
foundation_models
)
aurora_db = amazonaurora.AmazonAuroraVectorStore.from_existing_aurora_vector_store(
self, 'ExistingAuroraVectorStore',
cluster_identifier='aurora-serverless-vector-cluster',
database_name='bedrock_vector_db',
schema_name='bedrock_integration',
table_name='bedrock_kb',
vector_field='embedding',
text_field='chunks',
metadata_field='metadata',
primary_key_field='id',
embeddings_model=foundation_models.BedrockFoundationModel.COHERE_EMBED_ENGLISH_V3,
vpc=ec2.Vpc.from_lookup(self, 'VPC', vpc_id='vpc-0c1a234567ee8bc90'),
aurora_security_group_id='sg-012ef345678c98a76',,
secret=rds.DatabaseSecret.from_secret_complete_arn(
self,
'Secret',
Stack.of(self).format_arn(
service= 'secretsmanager',
resource= 'secret',
resource_name= 'rds-db-credentials/cluster-1234567890',
region= Stack.of(self).region,
account= Stack.of(self).account,
arn_format= ArnFormat.COLON_RESOURCE_NAME
)
)
)
kb = bedrock.KnowledgeBase(self, 'KnowledgeBase',
embeddings_model= foundation_models.BedrockFoundationModel.TITAN_EMBED_TEXT_V1,
vector_store=aurora_db,
instruction= 'Use this knowledge base to answer questions about books. ' +
'It contains the full text of novels.'
)
docBucket = s3.Bucket(self, 'DockBucket')
bedrock.S3DataSource(self, 'DataSource',
bucket= docBucket,
knowledge_base=kb,
data_source_name='books',
chunking_strategy= bedrock.ChunkingStrategy.FIXED_SIZE,
max_tokens=500,
overlap_percentage=20
)