AWS Snow Family now supports ML inference at the edge with Amazon SageMaker Edge Manager and AWS IoT Greengrass

AWS Snow Family now supports AWS IoT Greengrass version 2.0 or higher, which makes it easier for you to build IoT solutions and run ML inference on Snowball Edge and Snowcone devices. With support for AWS IoT Greengrass on Snow, you can now use Amazon SageMaker Edge Manager to optimize, secure, monitor, and maintain ML models on fleets of Snowball Edge devices.

Prior to this launch, you had to either run AWS IoT Greengrass on custom hardware that was not fully integrated with AWS services, or run AWS IoT Greengrass in your own AMIs on AWS Snow devices and install SageMaker Edge Manager software agent to run on Snow devices. Now, you can run your AWS IoT Greengrass software on AWS Snow’s Amazon Linux 2 (AL2) AMI for AWS IoT Greengrass and connect your device to AWS IoT services in minutes. When connected to an AWS region, you can also use AWS IoT Greengrass and SageMaker Edge Manager to collect prediction data and send a sample of the data to the cloud for monitoring, labeling, and retraining so you can maintain model accuracy over time. ML model data can be viewed in the SageMaker Edge Manager dashboard which reports on the operation of deployed models.

To get started, you simply need to select AWS IoT Greengrass when you order a Snow device from the AWS Snow Family console. Once you receive the device, you can use AWS OpsHub for Snow Family to unlock the device, set up your AWS IoT Greengrass environment, and start building your IoT solution in minutes. This feature is available on all Snowball and Snowcone devices that are ordered on or after June 14, 2021.

For more information about using AWS IoT Greengrass and Amazon SageMaker Edge Manager on AWS Snow Family devices, see launch blog post or visit the AWS Snowball documentation, AWS Snowball product page, and AWS Region availability page. Log into the AWS Snow Family console to get started.