How to use fully managed Jupyter notebooks in Amazon SageMaker
summary
Overview of how Amazon SageMaker runs fully managed Jupyter notebooks with concrete features: one-click SageMaker Studio notebooks that spin up elastic EC2-backed kernels and incur cost only when the kernel runs, fast launch <2 minutes, persistent user storage on EFS persistent home, lifecycle scripts for bootstrapping, and one-click sharing via read-only snapshot links. Demo builds an XGBoost marketing notebook, attaches a custom R image from Amazon ECR, switches instance types on the fly, and the shared read-only view preserves outputs and works as shown.
AI-written summary · about
chapters (7)
- 0:00Introduction to SageMaker
- 3:16Notebook instances overview
- 7:14Studio notebooks introduction
- 14:09Studio notebook features
- 18:03Demo notebook instances
- 23:14Demo Studio environment
- 31:37Summary and resources
related talks
32mSpin up Jupyter notebooks at scale and enhance your productivity
20mAWS On Air ft. Amazon SageMaker Studio Lab
35mAccelerate deep learning in the cloud with custom ML environments
49mUse Amazon SageMaker to build high-quality ML models faster
34mDeploying a Machine Learning System That Only Runs Locally
19mUsing Amazon SageMaker to develop and deploy ML models
34mBuild, train and deploy Machine Learning models with Amazon SageMaker
52mGenerate ML predictions without writing any code