Update Amazon Bedrock Knowledge Base > Introduction to Update Amazon Bedrock Knowledge Base recipe

Introduction to Update Amazon Bedrock Knowledge Base recipe

The Update Amazon Bedrock Knowledge Base recipe is based on REST and SOAP APIs. Use the recipe to update the existing Amazon Bedrock knowledge base using Amazon S3 data source and provide notification status.

The process is triggered by an HTTP request with the required input parameters, that is, knowledge base ID, email address, and an additional parameter data source ID if it is present or the bucket name to create a new data source using the bucket name.

The process then creates a data source if the user doesn't provide a data source ID. The name of the data source is the same as the name of the bucket to be added. If the user provides data source ID, the process updates the vector database.

The process then starts an ingestion job splitting the dataset into chunks and upserting them into the vector database. The process checks the ingestion status every 30 seconds, by default. You can change the interval as required.

The process then sends an email notification when the processing of files in the bucket is complete or to report any errors encountered during the process.