From the course: Machine Learning with SageMaker by Pearson
Unlock this course with a free trial
Join today to access over 25,200 courses taught by industry experts.
SageMaker auto scaling demonstration - Amazon SageMaker Tutorial
From the course: Machine Learning with SageMaker by Pearson
SageMaker auto scaling demonstration
Autoscaling in SageMaker will scale the number of instances running behind an endpoint. Specify a autoscaling policy, then you apply that to your variant on an endpoint, and then it will automatically scale the number of instances as required. I have this link in the lab documentation I have this link in the lab documentation. that takes you to the introduction to auto-scaling within SageMaker. There's several different ways to configure auto-scaling. You have the SDK, you have the command line interface, and you also have the web console. You can also do it with CloudFormation, as you can see over here, use AWS CloudFormation to create a scaling policy. We are going to do it with the web interface, the web console. I have a notebook in our lab directory that simply deploys a pre-trained model. The one unique thing that I'm doing here is giving it a static name so that I can then invoke or execute this predict.py application several times. So you can see I have six different windows…
Contents
-
-
-
-
-
-
-
-
-
-
(Locked)
Module introduction34s
-
(Locked)
Learning objectives33s
-
(Locked)
Using SageMaker Model Monitor for data drift and quality5m 49s
-
(Locked)
SageMaker Model Monitor demonstration6m 30s
-
(Locked)
Setting up alerts and CloudWatch dashboards7m 41s
-
(Locked)
Cost optimization with auto-scaling and SageMaker Savings Plans7m 9s
-
(Locked)
SageMaker auto scaling demonstration9m 31s
-
(Locked)
-
-