Amazon SageMaker — Survey of Ways to (Re)deploy Models on AWS
summary
Overview of Amazon SageMaker deployment options: Batch Transform, Asynchronous Inference, Serverless Inference, and Real-time Inference, with guidance on when to use each. Covers safe (re)deployment techniques including Variants and Deployment Guardrails, A/B testing and zero-downtime strategies, and touches on Multi-model and Multi-container endpoints for serving many models cost-effectively.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 8:29SageMaker overview
- 10:34Real-time inference
- 15:08Batch transform
- 20:45Asynchronous inference
- 23:12Serverless inference
- 28:27Model redeployment
- 31:18Blue-Green Guardrails
- 35:53Multi-variant testing
- 38:45Multi-model endpoints
- 45:05Deployment limits
- 47:12AWS service limits
related talks
45mDeploying ML Models to Production with Amazon SageMaker
01h 00mHigh-performance & cost-effective model deployment with Amazon SageMaker
33mBuild, train, and deploy ML models using Amazon SageMaker
32mBuild, Train, and Deploy Machine Learning Models on AWS
34mBuild, train and deploy Machine Learning models with Amazon SageMaker
19mUsing Amazon SageMaker to develop and deploy ML models
21mAWS On Air ft. Amazon SageMaker Inference Recommender
34mDeploying a Machine Learning System That Only Runs Locally