Exam Professional Cloud Architect All Questions

View all questions & answers for the Professional Cloud Architect exam

Exam Professional Cloud Architect topic 6 question 2 discussion

Actual exam question from Google's Professional Cloud Architect

Question #: 2
Topic #: 6

[All Professional Cloud Architect Questions]

For this question, refer to the Mountkirk Games case study. You need to analyze and define the technical architecture for the compute workloads for your company, Mountkirk Games. Considering the Mountkirk Games business and technical requirements, what should you do?

A. Create network load balancers. Use preemptible Compute Engine instances.
B. Create network load balancers. Use non-preemptible Compute Engine instances.
C. Create a global load balancer with managed instance groups and autoscaling policies. Use preemptible Compute Engine instances.
D. Create a global load balancer with managed instance groups and autoscaling policies. Use non-preemptible Compute Engine instances.

Show Suggested Answer

Suggested Answer: D 🗳️

by dabrat at Nov. 18, 2019, 10:16 p.m.

Comments

Submit Cancel

dabrat

Highly Voted 4 years, 2 months ago

D) => KPI game stability = Use non-preemptible

upvoted 50 times

tartar

3 years, 5 months ago

D is ok

upvoted 13 times

...

nitinz

2 years, 11 months ago

has to be C, A & B does not meet SLA. D does not meet KPI.

upvoted 2 times

...

KNG

Highly Voted 4 years ago

Agree "D". Preemptible VM is suitable for app which is fault-tolerant. Termination of preemptive VM might affect gaming experience, so it is not a good choice.

upvoted 21 times

...

Peto12

Most Recent 7 months, 1 week ago

Selected Answer: C

In the business requirements in the latest case study it's says at the bottom to minimise costs. Because of this I choose C.

upvoted 1 times

...

someCloudUser

11 months, 2 weeks ago

Selected Answer: D

D is correct in my opinion.

upvoted 1 times

...

megumin

1 year, 3 months ago

Selected Answer: D

ok for D

upvoted 1 times

...

AzureDP900

1 year, 3 months ago

D is right

upvoted 1 times

...

alexandercamachop

1 year, 4 months ago

"As the system scales, ensure that data is not lost due to processing backlogs" Answer is clearly D. Never mentions about cost saving / optimizing. Plus ending instances will affect current active users.

upvoted 3 times

...

Nirca

1 year, 4 months ago

Selected Answer: D

"D". should be the right one. Preemptible VM is suitable for app which is fault-tolerant. Preemptive might lead to availability or services issues in ONLINE applications

upvoted 1 times

...

Nirca

1 year, 6 months ago

Selected Answer: D

D it is!

upvoted 1 times

...

rsh3

2 years, 1 month ago

I got the point that Answer should include non-preemptible option, but I am confused why not B?

upvoted 1 times

panqueca

1 year, 8 months ago

They want to be everywhere around the globe, so u need a global load balancer to do this

upvoted 1 times

...

vincy2202

2 years, 1 month ago

Selected Answer: D

D is the correct answer

upvoted 1 times

...

joe2211

2 years, 2 months ago

Selected Answer: D

vote D

upvoted 2 times

...

MaxNRG

2 years, 3 months ago

D – create globale LB with MIG and autoscaling policies. Use non-preemptible Compute Engine instances. Standard vs Preemtible VMs is interesting dilemma for MountKirk Games. The diff in price is 4.5 times. E.g. 1 year price of n1-standard-8 (30 GB, 8 vCPUs) = 3329 $, and same preemptible = 740 $. If there are 10 servers running all time then saving in cost are 25890 $ per year. Likely not super extra expense for company if they want preserve and extend users. So, their choice should depend on user experience == their income. Couple notes from Case Study reqs supporting UX / standard VMs: 1) Our investors want more key performance indicators (KPIs) to evaluate the speed and stability of the game 2) Improve uptime - downtime is loss of players 3) Reduce latency to all customers 4) They plan to deploy the game’s backend on Google Compute Engine so they can capture streaming metrics, run intensive analytics, and take advantage of its autoscaling server environment and integrate with a managed NoSQL database.

upvoted 3 times

MaxNRG

2 years, 3 months ago

Last item about intestive analytics and streaming - is not appropriate task for preemptible VMs. And this is from GCP preemptible page: 1) If your applications are fault-tolerant and can withstand possible instance preemptions, then preemptible instances can reduce your Compute Engine costs significantly. For example, batch processing jobs can run on preemptible instances. 2) Preemptible instances cannot live migrate to a regular VM instance, or be set to automatically restart when there is a maintenance event. 3) Due to the above limitations, preemptible instances are not covered by any Service Level Agreement (and, for clarity, are excluded from the Google Compute Engine SLA). So, sounds as non-preemptible (standard) is a priority for Mountkirk...

upvoted 1 times

...