Exam AWS Certified Database - Specialty All Questions

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

Exam AWS Certified Database - Specialty topic 1 question 34 discussion

Exam question from Amazon's AWS Certified Database - Specialty

Question #: 34
Topic #: 1

[All AWS Certified Database - Specialty Questions]

A company is using an Amazon Aurora PostgreSQL DB cluster with an xlarge primary instance master and two large Aurora Replicas for high availability and read-only workload scaling. A failover event occurs and application performance is poor for several minutes. During this time, application servers in all Availability
Zones are healthy and responding normally.
What should the company do to eliminate this application performance issue?

A. Configure both of the Aurora Replicas to the same instance class as the primary DB instance. Enable cache coherence on the DB cluster, set the primary DB instance failover priority to tier-0, and assign a failover priority of tier-1 to the replicas.
B. Deploy an AWS Lambda function that calls the DescribeDBInstances action to establish which instance has failed, and then use the PromoteReadReplica operation to promote one Aurora Replica to be the primary DB instance. Configure an Amazon RDS event subscription to send a notification to an Amazon SNS topic to which the Lambda function is subscribed.
C. Configure one Aurora Replica to have the same instance class as the primary DB instance. Implement Aurora PostgreSQL DB cluster cache management. Set the failover priority to tier-0 for the primary DB instance and one replica with the same instance class. Set the failover priority to tier-1 for the other replicas.
D. Configure both Aurora Replicas to have the same instance class as the primary DB instance. Implement Aurora PostgreSQL DB cluster cache management. Set the failover priority to tier-0 for the primary DB instance and to tier-1 for the replicas.

Show Suggested Answer

Suggested Answer: C 🗳️

by Mickysingh at July 14, 2020, 3:28 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

RBSK

Highly Voted 2 years, 7 months ago

Selected Answer: C

Tier-0 is a requirement for CCM to work. Also having more than 1 read replica with Tier-0 also disables CCM - https://docs.amazonaws.cn/en_us/AmazonRDS/latest/AuroraUserGuide/aurora_ccm_status.html This clearly eliminates Option-D

upvoted 5 times

...

IhorK

Most Recent 1 year, 12 months ago

Selected Answer: C

"Cluster cache management is active on an Aurora PostgreSQL DB cluster when the cluster has an Aurora Reader instance configured as follows: The Aurora Reader instance uses same DB instance class type and size as the cluster's Writer instance. The Aurora Reader instance is configured as Tier-0 for the cluster. If the cluster has more than one Reader, this is its only Tier-0 Reader. Setting more than one Reader to Tier-0 disables CCM." https://docs.amazonaws.cn/en_us/AmazonRDS/latest/AuroraUserGuide/aurora_ccm_status.html

upvoted 1 times

jitesh_k

1 year, 8 months ago

Why do we need to set promotion tier for writer instance? It is already being used and failover will occur when it fails. When writer instance fails, reader replica will be promoted to writer instance. The writer instance that failed is not going to be removed from cluster - correct?

upvoted 1 times

...

sayed

2 years, 9 months ago

Selected Answer: C

C for CCM to work 1- you need to have one RR with the same instance class type and size as the writer instance 2- you need to set promotion priority to 0 for both writer and RR you need the details here https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraPostgreSQL.cluster-cache-mgmt.html

upvoted 4 times

...

sachin

3 years, 1 month ago

C is correct. https://aws.amazon.com/blogs/database/introduction-to-aurora-postgresql-cluster-cache-management/

upvoted 1 times

...

novice_expert

3 years, 3 months ago

Selected Answer: C

One Aurora Replica of same instance class as the primary DB -> Aurora PostgreSQL DB cluster cache management -> failover priority to tier-0 for the primary DB instance and one replica with the same instance class -> failover priority to tier-1 for the other replicas

upvoted 2 times

...

RotterDam

3 years, 4 months ago

Selected Answer: C

C is correct. Cluster Cache Management needs these three: - Set One Replica to have same priority as Primary - Set its instance class same as Primary

upvoted 4 times

...

tugboat

3 years, 5 months ago

C is economical, but priorities don't need to be set as largest replica will always be first target option for failovers

upvoted 2 times

...

VPup

3 years, 5 months ago

Selected Answer: C

CCM can be set to only one replica. That's why we put it to tier-0 to guarantee the fail over to the CCM enabled replica.

upvoted 4 times

...

thelad

3 years, 6 months ago

Answer is D. The replicas must be the same size as the primary when cluster cache management is used. https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraPostgreSQL.cluster-cache-mgmt.html

upvoted 4 times

novice_expert

3 years, 3 months ago

No all replicas but one where you want to fail over to, so ans is C

upvoted 3 times

...

shuraosipov

3 years, 8 months ago

Selected Answer: C

Answer is C. https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraPostgreSQL.cluster-cache-mgmt.html

upvoted 2 times

thelad

3 years, 6 months ago

From that link - "Cluster cache management requires that the designated reader instance have the same instance class type and size (db.r5.2xlarge or db.r5.xlarge, for example) as the writer" Therefore both replicas are required to be the same size as the primary. I'm going for Answer D

upvoted 1 times

novice_expert

3 years, 3 months ago

not both, just one is good

upvoted 1 times

...

GMartinelli

3 years, 8 months ago

Selected Answer: C

Option C

upvoted 2 times

...

Dip11

3 years, 9 months ago

Ans is C. Because this doc https://aws.amazon.com/blogs/database/introduction-to-aurora-postgresql-cluster-cache-management/ says to set priority of primary and one replica to zero. Which is not the case with D.

upvoted 2 times

...

shantest1

3 years, 9 months ago

C Answer Tier Priority 0 is the clue to eliminate D:

upvoted 2 times

anon9002

3 years, 9 months ago

The priority of 0 has nothing to do with it since more than one replica can have the same priority. The only difference appears to be cost and C will be cheaper. "You can customize the order in which your Aurora Replicas are promoted to the primary instance after a failure by assigning each replica a priority. Priorities range from 0 for the first priority to 15 for the last priority. If the primary instance fails, Amazon RDS promotes the Aurora Replica with the better priority to the new primary instance. You can modify the priority of an Aurora Replica at any time. Modifying the priority doesn't trigger a failover. More than one Aurora Replica can share the same priority, resulting in promotion tiers. If two or more Aurora Replicas share the same priority, then Amazon RDS promotes the replica that is largest in size. If two or more Aurora Replicas share the same priority and size, then Amazon RDS promotes an arbitrary replica in the same promotion tier. "

upvoted 2 times

im_not_robot

2 years, 5 months ago

CCM requires one replica has tier 0.

upvoted 1 times

...

LMax

3 years, 9 months ago

C and D look equally good to address the problem, but D would cost more as you upgrade all read replicas, not just 1. So for cost saving reasons would go with Answer C.

upvoted 1 times

...

Windy

3 years, 9 months ago

C for me.

upvoted 1 times

...

myutran

3 years, 9 months ago

Ans: C

upvoted 1 times

...

JobinAkaJoe

3 years, 9 months ago

I am torn between C &D. Performance issue after failover lasts only for a short duration which means it has more to do with cache management than instance sizing. CCM definitely is the solution. Between C & D, I will go with C considering the fact that originally read-replicas were of lower configuration for cost-saving, so its ideal to have one read-replica matching primary instance size and others with lower configuration if cost is a concern.

upvoted 1 times

...

Load full discussion...