Skip to content

Evaluating current workload using cloudwatch metrices

0

I want to find the cost estimate for MSK serverless for the current workload i have in my production environment. I cant find the MSK Serverless option in aws calculator, but in the aws msk pricing page Pricing Dimension Unit Price per unit Cluster-hours per hour, Partition-hours per hour, Storage per GiB-month, Data In per GiB, Data Out per GiB are there,

From the cloudwatch metrices how can i get the values for above dimension for one month. want to what cloudwatch metrices i should use and how

1 Answer
1
Accepted Answer

Here are the CloudWatch metrics you should monitor and how to use them to get the values for a one-month period:

1. Cluster-hours

You can monitor the uptime of your MSK clusters using the AWS/Kafka namespace in CloudWatch.

  • Metric: ActiveControllerCount
  • Statistic: Sum
  • Period: 1 Hour
  • Calculate the total hours by summing up the ActiveControllerCount metric over the desired period.

2. Partition-hours

To calculate partition-hours, you need to monitor the number of partitions in your MSK cluster over time.

  • Metric: PartitionsCount
  • Statistic: Sum
  • Period: 1 Hour
  • Sum up the PartitionsCount over the period to get the total partition-hours.

3. Storage per GiB-month

Monitor the storage used by your MSK cluster using the DataStorage metric.

  • Metric: DataStorage
  • Statistic: Maximum
  • Period: 1 Hour
  • Convert the value to GiB-month by taking the maximum value for each hour, summing it up over the month, and converting to GiB.

4. Data In per GiB

Monitor the amount of data ingested into your MSK cluster.

  • Metric: BytesInPerSec
  • Statistic: Sum
  • Period: 1 Hour
  • Convert the sum to GiB for the entire month.

5. Data Out per GiB

Monitor the amount of data consumed from your MSK cluster.

  • Metric: BytesOutPerSec
  • Statistic: Sum
  • Period: 1 Hour
  • Convert the sum to GiB for the entire month.
EXPERT
answered a year ago
EXPERT
reviewed a year ago
  • Thank you, additional small thing to ask, How do you determine these, is there any guide from aws side or is it from the experience and deep understanding of the services.

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.