Home icon

Embracing event driven architecture to enhance resilience of data solutions built on Amazon SageMaker

Big Data Blog

This article discusses strategies for enhancing the resilience of data solutions built on Amazon SageMaker through event-driven architecture. The key focus is on developing robust disaster recovery and backup strategies for data governance solutions.

  • Key design principles for resilience include:
    • Using serverless services
    • Documenting system metadata
    • Implementing comprehensive system health monitoring
  • Two backup patterns are presented:
    • Point-in-time backup (low RPO, seconds to minutes)
    • Scheduled backup (RPO in hours)
  • The solution uses AWS services like:
    • AWS Lambda
    • Amazon EventBridge
    • Amazon DynamoDB global tables
    • AWS Step Functions

The article provides detailed walkthrough guides for implementing backup strategies for both SageMaker Unified Studio and Amazon DataZone domains, with code samples and deployment instructions using AWS CDK.

Go to article

The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.

Related articles

Jun 30
2025
Better together: Amazon RDS for SQL Server and Amazon SageMaker Lakehouse, a generative AI data integration use case
Jun 6
2025
Simplify real-time analytics with zero-ETL from Amazon DynamoDB to Amazon SageMaker Lakehouse
May 3
2025
Building Resilient Distributed Systems with Temporal and AWS
Aug 5
2025
Integrate scientific data management and analytics with the next generation of Amazon SageMaker, Part 1

The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.