Exam DP-100 topic 4 question 2 discussion

Actual exam question from Microsoft's DP-100

Question #: 2
Topic #: 4

You create a deep learning model for image recognition on Azure Machine Learning service using GPU-based training.
You must deploy the model to a context that allows for real-time GPU-based inferencing.
You need to configure compute resources for model inferencing.
Which compute type should you use?

A. Azure Container Instance
B. Azure Kubernetes Service
C. Field Programmable Gate Array
D. Machine Learning Compute

Show Suggested Answer

Suggested Answer: B 🗳️
You can use Azure Machine Learning to deploy a GPU-enabled model as a web service. Deploying a model on Azure Kubernetes Service (AKS) is one option.
The AKS cluster provides a GPU resource that is used by the model for inference.
Inference, or model scoring, is the phase where the deployed model is used to make predictions. Using GPUs instead of CPUs offers performance advantages on highly parallelizable computation.
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/how-to-deploy-inferencing-gpus

by ACSC at March 27, 2021, 12:39 p.m.

Comments

Submit Cancel

ACSC

Highly Voted 4 years, 4 months ago

Answer is correct. AKS -> GPU support

upvoted 19 times

...

D0ktor

Most Recent 8 months, 4 weeks ago

Selected Answer: B

AKS is the solution

upvoted 1 times

...

Ran2025

1 year, 10 months ago

B is correct! AKS supports real-time GPU-based inferencing https://learn.microsoft.com/en-gb/azure/machine-learning/concept-compute-target?view=azureml-api-2

upvoted 1 times

...

phdykd

2 years ago

B. Azure Kubernetes Service Azure Kubernetes Service (AKS) provides an option to create a GPU-enabled node pool which would be suitable for real-time GPU-based inferencing. The other services listed do not provide the same level of GPU support necessary for such operations. Keep in mind that setting up and managing a Kubernetes cluster does require some additional skills and setup compared to other Azure services, but it provides a high level of control and scalability.

upvoted 1 times

...

phdykd

2 years, 5 months ago

The compute type that should be used for real-time GPU-based inferencing of a deep learning model is: D. Machine Learning Compute Explanation: To enable GPU-based inferencing on a deployed model, Machine Learning Compute (MLC) should be used. MLC is a managed service provided by Azure Machine Learning that can provision compute resources for training and inferencing machine learning models. The service can be configured to allocate resources based on the required processing power, which can include GPU and CPU-based clusters. Azure Container Instance (A) is a compute service that allows for running containerized applications without managing the underlying infrastructure, but it does not provide GPU resources. Azure Kubernetes Service (B) is a container orchestration service that can be used to manage containerized applications but requires additional configuration to enable GPU-based inferencing. Field Programmable Gate Array (C) is a hardware device that can be used to implement specific logic circuits but is not a cloud-based compute resource.

upvoted 3 times

...

austin06112000

3 years, 6 months ago

Answer is correct.

upvoted 1 times

...

dwight55

4 years ago

looking for somebody to share contrib access pdf file with Q&A & discussion m a i l to me at dwight (at) existiert.net ofc does not have to be free

upvoted 1 times

...

treadst0ne

4 years, 1 month ago

Answer is B. GPU for inference when deployed as a web service is supported only on AKS. https://docs.microsoft.com/en-gb/azure/machine-learning/concept-compute-target

upvoted 3 times

...