![]() This interval is the duration of the canary baking period. The parameter tells SageMaker to waitįor the specified amount of time (in seconds) between each interval shift. Note that the canary size shouldīe equal to or less than 50% of the green fleet's capacity.įor WaitIntervalInSeconds, use 600. The green fleet’s capacity as the canary. Percentage of your green fleet you want to use as the canary, and then set ![]() In the CanarySize field, you can change the size of the canary by modifying the This specifies that the deployment uses canary traffic shifting. TrafficRoutingConfiguration, set the Type parameter toĬANARY. Under DeploymentConfig and BlueGreenUpdatePolicy, in Update-endpoint command in the AWS CLI to initiate the deployment.įor EndpointName, use the name of the existing endpoint you want to update.įor EndpointConfigName, use the name of the endpoint configuration you want to use. You can use either the Amazon SageMaker UpdateEndpoint API or the ![]() Once you are ready for your deployment and have set up Amazon CloudWatch alarms for your endpoint, To learn more about CloudWatch alarms, see UsingĪmazon CloudWatch alarms in the Amazon CloudWatch User Guide. To learn how to set up CloudWatchĪlarms on an endpoint, see the prerequisite page Auto-Rollback Configuration and The alarms are active during the baking period, and if anyĪlarms trip, then all endpoint traffic rolls back to the blue fleet. Prerequisitesīefore setting up a deployment with canary traffic shifting, you must create Amazon CloudWatch alarms The green fleet serves all traffic and SageMaker terminates the blue fleet. If the final baking period finishes without tripping any alarms, then None of the alarms trip, then all of the traffic shifts to the green fleet and there is aįinal baking period. If any of the alarms trip during theīaking period, then SageMaker initiates a rollback and all traffic returns to the blue fleet. During this time, both the blue fleet and greenįleet are partially active and receiving traffic. Monitor the performance of the green fleet. Then the baking period begins, during which your CloudWatch alarms Once SageMaker provisions the green fleet, SageMaker routes a portion of the incoming traffic (forĮxample, 25%) to the canary. The following diagram shows how canary traffic shifting manages the distribution of traffic Canary traffic shifting provides you with more safety during yourĭeployment since any issues with the updated model only impact the canary. Pre-specified Amazon CloudWatch alarms trip, the rest of the traffic shifts from the old (blue) fleet Note that the canary size should be less than orĮqual to 50% of the new fleet's capacity. The portion of your green fleet that turns on to receive traffic is called the canary, and Lets you ensure that your new (green) fleet can serve inference before letting it handle 100% of You still have the benefits of a blue/green deployment, and the added canary feature Guardrail that validates the new fleet’s functionality before shifting all of your traffic to the With canary traffic shifting, you can test a portion of your endpoint traffic on the newįleet while the old fleet serves the remainder of the traffic.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |