AWS Certified AI PractitionerAWS Services for AI/ML and Generative AIHard
A global automotive manufacturer is developing an advanced driver-assistance system (ADAS) that uses machine learning models to analyze real-time video streams from vehicle cameras. These models need to be deployed directly onto the vehicle's embedded systems (edge devices) to perform inference with extremely low latency, even when there is no internet connectivity. The manufacturer also needs a way to efficiently update these models and manage the ML lifecycle on thousands of devices. Which AWS service is designed to facilitate the deployment and management of ML models on edge devices?
- AAmazon SageMaker
- BAmazon Rekognition Video
- CAWS IoT Greengrass
- DAWS Outposts
Show answer & explanationAnswer & explanation
Correct answer: C. AWS IoT Greengrass
AWS IoT Greengrass extends AWS cloud capabilities to edge devices, allowing them to perform ML inference locally. It enables deploying, running, and managing ML models on vehicles, ensuring low-latency processing even without internet connectivity. It also provides mechanisms for over-the-air (OTA) updates and lifecycle management for models on thousands of devices, directly meeting the requirements.
Why the other options are wrong
- A. Amazon SageMaker is for building, training, and deploying ML models in the cloud, but not directly for managing their deployment and inference on disconnected edge devices.
- B. Amazon Rekognition Video is a cloud-based computer vision service for video analysis; it is not designed for edge deployment and disconnected operations.
- D. AWS Outposts extends AWS infrastructure to on-premises data centers, but it's for larger-scale on-premises deployments, not for managing ML models on thousands of individual, potentially disconnected, edge devices like vehicle embedded systems.
AWS IoT Greengrass
An IoT edge runtime and cloud service that helps you build, deploy, and manage intelligent device software.
- Extends AWS capabilities (e.g., Lambda, ML inference) to edge devices.
- Enables local processing, even offline.
- Supports over-the-air (OTA) updates for applications and ML models.
Memory trick: Greengrass brings the cloud to the edge.