Is 'cpu_usage_active' calculated by CloudWatch agent same as CPUUtilization under AWS/ECS namespace?
I am currently trying to create a custom metric based on CPUUtilization for faster autoscaling. The one provided by AWS with 1-min resolution is not good enough. While finding a solution on how to create one, I came across this blog post which shows using System Manager and unified Cloud Agent to create a scaling policy using Memory Utilization - BLOG.
I checked and found that there is no metric collected by the CloudWatch Agent under the name CPUUtilization but there is one under the name cpu_usage_active. So, I am confused as to whether both the metrics are same or different. If different, then how so?
Amazon ECS supports CPU utilization metrics across multiple dimensions: cluster and service.
Cluster-wide average CPU utilization is calculated by adding the CPU resources consumed by all tasks running on EC2 instances in the cluster and dividing it by the total CPU resources among all EC2 instances registered with the cluster. For example, if your total EC2 capacity in your ECS cluster is 100 vCPUs, and your tasks are consuming a total of 50 vCPUs, then the average cluster-wide CPU utilization will be 0.5 (50%). This value should be similar to the average value reported by a collection of on-instance agents (such as the Amazon CloudWatch Agent) running on the same EC2 instances. Cluster CPU utilization is only applicable to EC2 instances; Fargate tasks will not be included in this calculation.
Service-wide average CPU utilization is calculated by adding the CPU utilization of all tasks running in the service and dividing it by the total CPU reservation of all tasks running in the service. For example, if you have 100 tasks running, each task has a reservation of 1024 CPU units (1 vCPU), and the tasks are currently consuming 40960 CPU units (40 vCPUs) in total, then the average CPU utilization will be 40%. The total vCPU capacity of the EC2 instances isn't considered for service-wide CPU utilization. And service-wide CPU utilization includes Fargate task utilization.
Manual remediation config works, automatic remediation config failsasked 12 days ago
Is 'cpu_usage_active' calculated by CloudWatch agent same as CPUUtilization under AWS/ECS namespace?asked 3 months ago
how can I configure cloudwatch to monitor things like CPU/RAM/Disk/IO for auto scaling groups using a wildcard?asked 5 months ago
CloudWatch Agent Metric Namespaces not respecting configurationasked 2 years ago
Autoscaling with pythonasked 2 months ago
Fetch multiple couldwatch metrics in one cli callAccepted Answerasked a month ago
CloudWatch stopping tail when file is deletedasked 2 years ago
How to receive only one sns notification triggered via autoscaling event.Accepted Answerasked 18 days ago
Reboot an instance, with a Cloud Watch Alarms based on custom metricasked 2 years ago
Cloudwatch Fees from Metric Storage Issueasked 2 years ago