exam questions

Exam AWS Certified Solutions Architect - Professional All Questions

View all questions & answers for the AWS Certified Solutions Architect - Professional exam

Exam AWS Certified Solutions Architect - Professional topic 1 question 623 discussion

A financial company is using a high-performance compute cluster running on Amazon EC2 instances to perform market simulations. A DNS record must be created in an Amazon Route 53 private hosted zone when instances start. The DNS record must be removed after instances are terminated.
Currently the company uses a combination of Amazon CloudWatch Events and AWS Lambda to create the DNS record. The solution worked well in testing with small clusters, but in production with clusters containing thousands of instances the company sees the following error in the Lambda logs:
HTTP 400 error (Bad request).
The response header also includes a status code element with a value of `Throttling` and a status message element with a value of `Rate exceeded`.
Which combination of steps should the Solutions Architect take to resolve these issues? (Choose three.)

  • A. Configure an Amazon SOS FIFO queue and configure a CloudWatch Events rule to use this queue as a target. Remove the Lambda target from the CloudWatch Events rule.
  • B. Configure an Amazon Kinesis data stream and configure a CloudWatch Events rule to use this queue as a target. Remove the Lambda target from the CloudWatch Events rule.
  • C. Update the CloudWatch Events rule to trigger on Amazon EC2 ג€Instance Launch Successfulג€ and ג€Instance Terminate Successfulג€ events for the Auto Scaling group used by the cluster.
  • D. Configure a Lambda function to retrieve messages from an Amazon SQS queue. Modify the Lambda function to retrieve a maximum of 10 messages then batch the messages by Amazon Route 53 API call type and submit. Delete the messages from the SQS queue after successful API calls.
  • E. Configure an Amazon SQS standard queue and configure the existing CloudWatch Events rule to use this queue as a target. Remove the Lambda target from the CloudWatch Events rule.
  • F. Configure a Lambda function to read data from the Amazon Kinesis data stream and configure the batch window to 5 minutes. Modify the function to make a single API call to Amazon Route 53 with all records read from the kinesis data stream.
Show Suggested Answer Hide Answer
Suggested Answer: ACD 🗳️

Comments

Chosen Answer:
This is a voting comment (?). It is better to Upvote an existing comment if you don't have anything to add.
Switch to a voting comment New
bbnbnuyh
Highly Voted 3 years, 7 months ago
C, D, E You have to introduce a SQS: FIFO has limited throughput so may be a normal SQS queue with batching that can overcome the rate limits
upvoted 28 times
beso
3 years, 7 months ago
CloudWatch--> SQS--> Lambda (batch) --> R53
upvoted 3 times
...
Kelvin
3 years, 7 months ago
ACD is better as you need FIFO mode to ensure processing DNS records exactly once.
upvoted 6 times
...
pablobairat
3 years, 6 months ago
ACD If you use UPSERT to introduce the DNS records, if it does not exists, it creates it, if it exists, it update the values (in case of duplicates). For deleting, you use a delete, if it exists, it deletes it, if it was already deleted(duplicated message in the queue), it does nothing. The goal here is to support thousands of instances launching and terminating, with a SQS FIFO queue this requirement is not fullfilled. And it was the original problem with Lambda and the concurrency.
upvoted 6 times
...
...
Ebi
Highly Voted 3 years, 7 months ago
I will go with ACD
upvoted 13 times
Ebi
3 years, 7 months ago
We need FIFO queue here for exactly-once-processing feature as well as order
upvoted 4 times
...
LCC92
3 years, 6 months ago
FIFO SQS is limited 300 message/second. CDE is correct.
upvoted 5 times
...
...
Jesuisleon
Most Recent 1 year, 11 months ago
Selected Answer: ACD
A,C,D. Here the problem is not the speed that SQS FIFO is less than SQS, it's the order we need to maintain. Each instance should have a record in Route 53 when it's started and shoud have this record deleted from Route 53 when the instance is terminated. You don't want to record is deleted before its creation in your environment.
upvoted 1 times
...
maxh8086
2 years, 4 months ago
https://docs.aws.amazon.com/AmazonCloudWatch/latest/events/RelayEventsKinesisStream.html https://docs.aws.amazon.com/AmazonCloudWatch/latest/events/CloudWatch-Events-tutorial-CloudWatch-Logs.html
upvoted 1 times
...
evargasbrz
2 years, 4 months ago
Selected Answer: CDE
CDE is the answer
upvoted 1 times
...
JohnPi
2 years, 7 months ago
Selected Answer: CDE
CDE is the answer
upvoted 2 times
...
aqiao
2 years, 8 months ago
Selected Answer: ACD
Even the default message groups support 300 requests , you can enable high performance option to improve high throughput:https://docs.aws.amazon.com/AWSSimpleQueueService/latest/SQSDeveloperGuide/high-throughput-fifo.html
upvoted 1 times
...
jj22222
3 years, 1 month ago
Selected Answer: ACD
ACD looks right
upvoted 1 times
...
jyrajan69
3 years, 2 months ago
If you are choosing E, then you are okay with duplicate DNS records, which means that when you delete the records you have to figure out the timestamp so that you do not delete the latest entry. Yes you do have limited throughput but thats where D comes in with batching, now you can get 3000 TPS for FIFO. So my answer will have to be A,C,D
upvoted 2 times
...
tkanmani76
3 years, 4 months ago
Answer C, D, E - If we would have gone with A, C, D - which makes sense from FIFO perspective, the option D does not mention 'SQS FIFO' instead just mentions SQS. Hence C, D, E
upvoted 1 times
...
AzureDP900
3 years, 5 months ago
CDE is perfect answer, This question in Neal Davis practice test. The errors in the Lambda logs indicate that throttling is occurring. Throttling is intended to protect your resources and downstream applications. Though Lambda automatically scales to accommodate incoming traffic, functions can still be throttled for various reasons. In this case it is most likely that the throttling is not occurring in Lambda itself but in API calls made to Amazon Route 53. In Route 53 you are limited (by default) to five requests per second per AWS account. If you submit more than five requests per second, Amazon Route 53 returns an HTTP 400 error (Bad request). The response header also includes a Code element with a value of Throttling and a Message element with a value of Rate exceeded. The resolution here is to place the data for the DNS records into an SQS queue where they can buffer. AWS Lambda can then poll the queue and process the messages, making sure to batch the messages to reduce the likelihood of receiving more errors.
upvoted 6 times
...
Cotter
3 years, 6 months ago
CDE better than ADE FIFO SQS is limited 300 message/second as commented below.
upvoted 1 times
...
andylogan
3 years, 6 months ago
It's C, D, E
upvoted 1 times
...
tgv
3 years, 6 months ago
CCC DDD EEE ---
upvoted 2 times
...
WhyIronMan
3 years, 6 months ago
I'll go with C,D,E
upvoted 2 times
...
Chibuzo1
3 years, 6 months ago
The Correct Answer is C D E.
upvoted 1 times
...
vkbajoria
3 years, 6 months ago
C, D , E
upvoted 1 times
...
Community vote distribution
A (35%)
C (25%)
B (20%)
Other
Most Voted
A voting comment increases the vote count for the chosen answer by one.

Upvoting a comment with a selected answer will also increase the vote count towards that answer by one. So if you see a comment that you already agree with, you can upvote it instead of posting a new comment.

SaveCancel
Loading ...
exam
Someone Bought Contributor Access for:
SY0-701
London, 1 minute ago