From the course: AWS Certified Machine Learning Engineer Associate (MLA-C01) Cert Prep
Unlock this course with a free trial
Join today to access over 24,700 courses taught by industry experts.
Amazon SageMaker Serverless Inference
From the course: AWS Certified Machine Learning Engineer Associate (MLA-C01) Cert Prep
Amazon SageMaker Serverless Inference
- [Instructor] Hello, guys, and welcome. So, in today's session, we are going to talk about Amazon SageMaker Serverless Inference. So, Amazon SageMaker Serverless Inference, it's actually a purpose-built inference which allows you to deploy machine learning models for inference without the need to either configure or manage the underlying infrastructure, and this totally simplifies the deployment process and reduces the operational overhead. It's ideal for workloads with idle periods, so it's actually ideal for workloads that have the idle periods between traffic bursts, and could tolerate the cold starts. So, for example, it's useful for applications with unpredictable traffic, such as certain web applications or batch processing tasks. So, it's actually good for scaling, so if there are no requests, then the serverless endpoint scales down to zero, making it a cost-effective solution, so you only pay for what you use, and this significantly reduces the costs during the idle periods,…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
-
-
-
(Locked)
Intro: Model deployment53s
-
(Locked)
Online inference (real-time)20m 57s
-
(Locked)
Batch transform2m 17s
-
(Locked)
Other deployments8m 8s
-
(Locked)
Multi-model vs. multi-container endpoints10m 24s
-
(Locked)
Hands-on learning: Multi-model endpoint7m 16s
-
Hands-on learning: Multi-container endpoint2m 49s
-
(Locked)
SageMaker deployment7m 48s
-
(Locked)
Hands-on learning: XGBoost (churn prediction)6m 43s
-
(Locked)
Hands-on learning: Script mode3m 1s
-
(Locked)
Hands-on learning: Bring your own (BYO) Docker4m
-
(Locked)
SageMaker instance types3m 2s
-
(Locked)
SageMaker SDK7m 11s
-
(Locked)
Distributed training5m 20s
-
(Locked)
SageMaker Debugger3m 33s
-
Hands-on learning: SageMaker serverless inference6m 9s
-
(Locked)
SageMaker Autopilot3m 33s
-
(Locked)
Amazon SageMaker Inference Recommender6m 37s
-
(Locked)
Amazon SageMaker Serverless Inference5m 24s
-
(Locked)
Inference pipeline5m 3s
-
(Locked)
Hands-on learning: SageMaker Model Monitor15m 51s
-
(Locked)
SageMaker Neo6m 29s
-
(Locked)
SageMaker security6m 54s
-
(Locked)
Deployment target services10m 10s
-
(Locked)
Maintainable, scalable, cost-effective deployments8m 38s
-
(Locked)
Automatic scaling metrics4m 16s
-
(Locked)
Performance tradeoff analysis4m 10s
-
(Locked)
Apache Airflow, SageMaker Pipelines6m
-
(Locked)
Isolated ML system13m 12s
-
(Locked)
Exam cram11m 16s
-
(Locked)
-
-
-
-