You have an Azure Machine Learning workspace. You build a deep learning model. You need to publish a GPU-enabled model as a web service. Which two compute targets can you use? Each correct answer pres

The correct answer is A. Azure Kubernetes Service (AKS) D. Azure Machine Learning compute clusters. Azure Kubernetes Service (AKS) supports GPU-enabled node pools and is the recommended compute target for deploying production-grade, GPU-accelerated model web services with low latency and high throughput. Azure Machine Learning compute clusters (option D) can also be provisioned

Train and deploy models

Question

You have an Azure Machine Learning workspace. You build a deep learning model. You need to publish a GPU-enabled model as a web service. Which two compute targets can you use? Each correct answer presents a complete solution. NOTE: Each correct selection is worth one point.

Options

AAzure Kubernetes Service (AKS)
BAzure Container Instances (ACI)
CLocal web service
DAzure Machine Learning compute clusters

How the community answered

(31 responses)

A
81% (25)
B
6% (2)
C
13% (4)

Explanation

Azure Kubernetes Service (AKS) supports GPU-enabled node pools and is the recommended compute target for deploying production-grade, GPU-accelerated model web services with low latency and high throughput. Azure Machine Learning compute clusters (option D) can also be provisioned with GPU VM sizes and support batch inference endpoints, making them valid targets for GPU-enabled model serving scenarios. Azure Container Instances (option B) has very limited and unreliable GPU support and is not recommended for GPU workloads. A local web service (option C) relies on the local machine's hardware and provides no managed GPU support for production deployments.

Topics

#Model Deployment#Azure ML Compute#GPU Inference#Web Services

Community Discussion

No community discussion yet for this question.

Full DP-100 Practice