Exam AWS Certified Machine Learning - Specialty All Questions

View all questions & answers for the AWS Certified Machine Learning - Specialty exam

Exam AWS Certified Machine Learning - Specialty topic 1 question 180 discussion

Exam question from Amazon's AWS Certified Machine Learning - Specialty

Question #: 180
Topic #: 1

[All AWS Certified Machine Learning - Specialty Questions]

A company has an ecommerce website with a product recommendation engine built in TensorFlow. The recommendation engine endpoint is hosted by Amazon
SageMaker. Three compute-optimized instances support the expected peak load of the website.
Response times on the product recommendation page are increasing at the beginning of each month. Some users are encountering errors. The website receives the majority of its traffic between 8 AM and 6 PM on weekdays in a single time zone.
Which of the following options are the MOST effective in solving the issue while keeping costs to a minimum? (Choose two.)

A. Configure the endpoint to use Amazon Elastic Inference (EI) accelerators.
B. Create a new endpoint configuration with two production variants.
C. Configure the endpoint to automatically scale with the InvocationsPerInstance metric.
D. Deploy a second instance pool to support a blue/green deployment of models.
E. Reconfigure the endpoint to use burstable instances.

Show Suggested Answer

Suggested Answer: AC 🗳️

by bluer1 at May 1, 2022, 8:08 p.m.

Disclaimers:

- ExamTopics website is not related to, affiliated with, endorsed or authorized by Amazon.
- Trademarks, certification & product names are used for reference only and belong to Amazon.

Comments

Submit Cancel

Sivadharan

Highly Voted 3 years, 1 month ago

Selected Answer: AC

AC - for me https://aws.amazon.com/machine-learning/elastic-inference/ https://aws.amazon.com/blogs/machine-learning/configuring-autoscaling-inference-endpoints-in-amazon-sagemaker/

upvoted 16 times

Jerry84

2 years, 4 months ago

Agree. The problem with E is that it mentioned "majority of its traffic between 8 AM and 6 PM on weekdays", therefore it is not cost effective during period from 6PM to 8AM weekdays and weekends. Whereas the auto-scaling(C) could save money during all the time.

upvoted 1 times

...

rohit07cf

Highly Voted 3 years, 1 month ago

AC looks correct

upvoted 9 times

...

sheetalconect

Most Recent 1 year ago

Selected Answer: CE

traffic between 8 AM and 6 PM on weekdays in a single time zone .--- Reconfigure the endpoint to use burstable instances. Configure the endpoint to automatically scale with the InvocationsPerInstance metric.

upvoted 2 times

...

loict

1 year, 9 months ago

Selected Answer: AC

A. YES - EI provides GPU access (Sept 2023: now deprecated) B. NO - not best practice to scale (altough it might help ?) C. YES - you want to scale with traffic D. NO - not best practice to scale (altough it might help ?) E. NO - burstable instances is for inpredictable traffic, bursting for long period of time is not cost effective

upvoted 2 times

...

Mickey321

1 year, 9 months ago

Selected Answer: CE

Either AC or CE very confusing but C is confirmed and will go with E

upvoted 2 times

Mickey321

1 year, 9 months ago

Or since the question is really old so maybe AC is what needs the answer to be

upvoted 1 times

...

Mickey321

1 year, 9 months ago

Selected Answer: AC

A C for me

upvoted 1 times

...

kaike_reis

1 year, 10 months ago

Selected Answer: AC

A - C for me E is costy

upvoted 1 times

...

ZSun

2 years, 1 month ago

E is not correct. A C E are both effective method to optimize inference, but A is used for GPU, and E is used for CPU. since this question is about Tensorflow, it should be A and E is not effective.

upvoted 1 times

...

blanco750

2 years, 3 months ago

Selected Answer: AC

A and C are the right answers

upvoted 3 times

...

luckybme

2 years, 3 months ago

Interesting if option A will be relevant anymore, as AWS is discontinuing Elastic Inference starting Apr 15. https://docs.aws.amazon.com/sagemaker/latest/dg/ei.html. Wonder if they change the option to include Inf instane type

upvoted 3 times

...

wjohnny

2 years, 6 months ago

Selected Answer: CE

Why elastic inference given that GPU is not necessary? I guess C and E (IMO)

upvoted 2 times

...

Ob1KN0B

2 years, 10 months ago

Selected Answer: AE

A - This is using Tensorflow which means Elastic inference can be used to save costs for GPU, thereby reducing the compute time. E - Since the load is not uniform, it will help to use burstable instances to operate above the threshold when the situation demands. C is not at all cost effective.

upvoted 3 times

...