Using Sagemaker with RDS Data | S02EP11 Lets talk about data show

Using Sagemaker with RDS Data | S02EP11 Lets talk about data show

In this show we would be discussing about Amazon Sagemaker and how it can be used for Machine learning and predictions of database objects.

Prasad Matkar
Amazon Employee
Published Mar 27, 2024
In this host builder episode of "Let's Talk About Data" show hosted by Tony and Prasad they demonstrate how to get data out of an RDS PostgreSQL database into an S3 bucket using a simple SQL query. They then catalog the data in AWS Glue and query it using Athena. Tony shows how to quickly visualize the data in Amazon QuickSight and adds basic forecasting.
Prasad then demonstrates using Amazon SageMaker to build a time series forecasting model to predict future database growth. He shows how SageMaker Canvas provides a no-code environment to import data, configure and train ML models, get predictions, and deploy them. In the show Prasad was able to show the SageMaker model forecasting future database size growth and visualizing the output of exported Sagemaker csv file using Amazon QuickSight.
Other key highlights include:
  • Using the S3 export capability built into RDS PostgreSQL to efficiently extract data
  • Cataloging data with AWS Glue crawlers to make visible for Athena queries
  • Visualizing data quickly with Amazon QuickSight and adding basic ML-powered forecasting
  • Building custom ML models with no-code using Amazon SageMaker Canvas
  • Leveraging SageMaker capabilities to forecast time series data like database growth
Prasad Matkar - Database Specialist Solutions Architect @ AWS
Tony Mullen - Senior RDS Specialist Solutions Architect @ AWS

  • Amazon SageMaker Canvas - https://aws.amazon.com/sagemaker/canvas/
  • Amazon Sagemaker Pricing - https://aws.amazon.com/sagemaker/canvas/pricing/
  • Getting started with Amazon SageMaker - https://catalog.us-east-1.prod.workshops.aws/workshops/0c6b8a23-b837-4e0f-b2e2-4a3ffd7d645b/en-US/getting-started/setup

Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.