Exam Professional Cloud Architect All Questions

View all questions & answers for the Professional Cloud Architect exam

Exam Professional Cloud Architect topic 8 question 1 discussion

Actual exam question from Google's Professional Cloud Architect

Question #: 1
Topic #: 8

[All Professional Cloud Architect Questions]

TerramEarth's CTO wants to use the raw data from connected vehicles to help identify approximately when a vehicle in the field will have a catastrophic failure.
You want to allow analysts to centrally query the vehicle data.
Which architecture should you recommend?
A.

B.

C.

D.

Show Suggested Answer

Suggested Answer: A
The push endpoint can be a load balancer.
A container cluster can be used.
Cloud Pub/Sub for Stream Analytics

Reference:
https://cloud.google.com/pubsub/
https://cloud.google.com/solutions/iot/
https://cloud.google.com/solutions/designing-connected-vehicle-platform https://cloud.google.com/solutions/designing-connected-vehicle-platform#data_ingestion http://www.eweek.com/big-data-and-analytics/google-touts-value-of-cloud-iot-core-for-analyzing-connected-car-data https://cloud.google.com/solutions/iot/

by MJK at Dec. 30, 2019, 12:17 p.m.

Comments

Submit Cancel

hems4all

Highly Voted 4 years, 3 months ago

A is correct As described in the Designing a Connected Vehicle Platform on Cloud IoT Core case study, 1. Google Cloud Dataflow is essential to transform, enrich and then store telemetry data by using distributed data pipelines 2. Cloud Pub/Sub is essential to handle the streams of vehicle data while at the same time decoupling the specifics of the backend processing implementation It now comes down to a choice between 1. Cloud SQL vs BigQuery for analytics. 2. GKE (with or without Anthos) + Cloud Load balancing vs App Engine. For the first point, there is no doubt that BigQuery is the preferred choice for analytics. Cloud SQL does not scale to this sort of data volume (9TB/day + data coming through when vehicles are serviced). For the second point, GKE with Cloud Load Balancing is a better fit than App Engine. App Engine is a regional service whereas, with the other option, you can have multiple GKE clusters in different regions. And Cloud Load Balancing can send requests to the cluster in the region that is closest to the vehicle. This option minimizes the latency and makes the feedback loop more real-time.

upvoted 63 times

...

MJK

Highly Voted 5 years, 1 month ago

Ans should be A

upvoted 13 times

nitinz

3 years, 11 months ago

It is A

upvoted 1 times

...

Jphix

4 years, 1 month ago

Agreed - A. For those with load balancing concerns re: FTP; transport protocol for ftp is TCP, albeit over two TCP ports. GCP Load Balancing supports TCP.

upvoted 3 times

...

GCP_Azure

4 years, 9 months ago

You will need to use (install) an FTP server that supports load balancing. Google Cloud does not offer an FTP server service or software product

upvoted 3 times

alii

4 years ago

App engine is regional https://cloud.google.com/appengine/docs/locations. Case study says it's a global business. which rules out app engine. so we are left with A. "App Engine is regional, which means the infrastructure that runs your apps is located in a specific region, and Google manages it so that it is available redundantly across all of the zones within that region."

upvoted 4 times

alii

4 years ago

As per case study: "They currently have over 500 dealers and service centers in 100 countries."

upvoted 1 times

...

tartar

4 years, 6 months ago

B is ok

upvoted 9 times

bnlcnd

4 years ago

B missed the global LB. App Engine is regional only. You have to create App Engine in each region and use LB as the front door.

upvoted 1 times

...

bjuneja

4 years, 2 months ago

How can you do analysis on SQL? A is ok

upvoted 5 times

...

salvo007

Most Recent 1 year ago

Load Balancer doesn't work on ftp. You need a server to close the connectione and put the file. than the server ftp, can put the file on pub sub etc. so the answer is B

upvoted 2 times

...

Jannchie

1 year, 2 months ago

App Engine is not for analytic task, so GKE is better. A.

upvoted 1 times

...

Jconnor

1 year, 2 months ago

yup. It's A. Maybe FTP in passive mode would help.

upvoted 1 times

...

rr4444

1 year, 11 months ago

This question is a mess FTP all over the place And even into things that can't talk FTP in that way Needs fixing

upvoted 4 times

...

BeCalm

1 year, 11 months ago

What is a Google Container Engine?

upvoted 1 times

n_nana

1 year, 11 months ago

It is GKE in old google naming

upvoted 3 times

...

Mahmoud_E

2 years, 3 months ago

A is correct

upvoted 1 times

...

JoeThach

2 years, 8 months ago

I vote for A - The company has customers in 100+ countries, while App Engine is regional (rule out B).

upvoted 1 times

...

amxexam

2 years, 9 months ago

B is correct. We dont have LB that support SFTP traffic. Hence we eleminate A & C We are talking arbore analyzing IoT data = Big Quavery = B.

upvoted 2 times

moota

2 years ago

Google Cloud Load Balancing can also do TCP/SSL load balancing. Check out key features in https://cloud.google.com/load-balancing

upvoted 1 times

HD2023

1 year, 10 months ago

"FTP" doesn’t appear once on that entire page. Answer: B

upvoted 1 times

...

jpco

2 years, 11 months ago

Google Load Balancer doesn't support FTP protocol

upvoted 3 times

...

joe2211

3 years, 2 months ago

vote A

upvoted 2 times

...

medeis_jar

3 years, 4 months ago

Answer B, because of Bigquery (Analytics) and AppEngine Flex (custom runtime environments) and the fact that there is no such a thing as Google Container Engine, there is Google Kubernetes Engine.

upvoted 3 times

...

Nik22

3 years, 4 months ago

These are from old use case. Do we still need to do these?

upvoted 5 times

...

MamthaSJ

3 years, 7 months ago

Answer is A

upvoted 2 times

...

XDevX

3 years, 7 months ago

IMHO it is b). Background: We have in 22 hours 9 TB of ingest data. That makes 113 MB per second. The question is: Why do we have to use for a "simple" transformation of FTP files into Cloud Pub/Sub GKE with multiple clusters around the world? For me it seems to be sufficient to have App Engine Flexible in one region to handle that. We have no highly interactive game where every ms counts concerning the customer experience - we have "only" some vehicles sending data that has to be tranformed and then analysed.

upvoted 4 times

...

gatul28

3 years, 8 months ago

how B? I do not see any LB there and App Engine doesn't support LBs beyond HTTP(s). Answer Is A but B's existence is confusing me much

upvoted 1 times

HD2023

1 year, 10 months ago

ultra low latency isn’t required. Whereas A isn’t possible because you can’t use FTP with LB.

upvoted 1 times

...

Load full discussion...