Microsoft DP-100 Exam - Topic 1 Question 31 Discussion

Actual exam question for Microsoft's DP-100 exam

Question #: 31
Topic #: 1

An organization creates and deploys a multi-class image classification deep learning model that uses a set of labeled photographs.

The software engineering team reports there is a heavy inferencing load for the prediction web services during the summer. The production web service for the model fails to meet demand despite having a fully-utilized compute cluster where the web service is deployed.

You need to improve performance of the image classification web service with minimal downtime and minimal administrative effort.

What should you advise the IT Operations team to do?

AIncrease the minimum node count of the compute cluster where the web service is deployed.

BCreate a new compute cluster by using larger VM sizes for the nodes, redeploy the web service to that cluster, and update the DNS registration for the service endpoint to point to the new cluster.

CIncrease the VM size of nodes in the compute cluster where the web service is deployed.

DIncrease the node count of the compute cluster where the web service is deployed.

Show Suggested Answer

Suggested Answer: D

The Azure Machine Learning SDK does not provide support scaling an AKS cluster. To scale the nodes in the cluster, use the UI for your AKS cluster in the Azure Machine Learning studio. You can only change the node count, not the VM size of the cluster.

https://docs.microsoft.com/en-us/azure/machine-learning/how-to-create-attach-kubernetes

by Helga at May 05, 2022, 09:44 PM

Limited Time Offer

25%

Off

Get Premium DP-100 Questions as Interactive Web-Based Practice Test or PDF

Contribute your Thoughts:

Submit Cancel

Teri

4 months ago

Not sure if just adding more nodes will solve the problem.

upvoted 0 times

...

Jenise

4 months ago

Wait, why not just optimize the existing model instead?

upvoted 0 times

...

Horace

4 months ago

Definitely agree with option C, bigger VMs can handle more load.

upvoted 0 times

...

Elinore

5 months ago

I think creating a new cluster with larger VMs is better for long-term performance.

upvoted 0 times

...

Ronnie

5 months ago

Increasing the node count is a quick fix.

upvoted 0 times

...

Clay

5 months ago

I feel like we practiced a similar question where increasing the node count was the right answer, but I can't recall if it was for a web service specifically.

upvoted 0 times

...

Flo

5 months ago

Increasing the VM size of the nodes seems like a straightforward solution, but I wonder if it would really handle the demand during peak times.

upvoted 0 times

...

Evangelina

5 months ago

I think creating a new compute cluster with larger VM sizes could help, but it sounds like it might involve more downtime than we want.

upvoted 0 times

...

Georgiana

5 months ago

I remember we discussed scaling options in class, but I'm not sure if just increasing the node count is enough for heavy loads.

upvoted 0 times

...

Mirta

5 months ago

I'm a little confused by this question. I know jailbreaking is modifying iOS to get past Apple's restrictions, but I'm not super familiar with the specific tools used. I'll have to make an educated guess on this one.

upvoted 0 times

...

Cassi

5 months ago

Hmm, I'm not sure about this one. The image doesn't give a lot of context, so I'll have to think it through carefully.

upvoted 0 times

...

Harrison

5 months ago

Okay, let me think this through. The question mentions that the model was published as a real-time service on an AKS inference compute cluster, so I'm guessing the name of the AKS cluster and the name of the inference pipeline would also be important details to share.

upvoted 0 times

...