In previous chapter you verified that the Auto Scaling group successfully forecasted with predictive scaling policy. As predictive scaling starts launching instances only beginning of each hour, to save time you can proceed with this next step.
With predictive scaling, EC2 Auto Scaling does not scale in when the predicted capacity is lower than the actual capacity. Due to this behavior, you need to combine predictive scaling with another scaling policy to scale in capacity when it’s not needed.
Dynamic scaling is used to automatically scale capacity in response to real-time changes in resource utilization. Using it with predictive scaling helps you follow the demand curve for your application closely, scaling in during periods of low traffic and scaling out when traffic is higher than expected to react to demand changes or spikes that were not forecasted by predictive scaling.
When multiple scaling policies are active, each policy determines the desired capacity independently, and the desired capacity is set to the maximum of those.
In this section, you configure the Auto Scaling group to automatically scale out and scale in as your application load fluctuates. You use a average CPU utilization across the Auto Scaling group instances.
cat <<EoF > asg-automatic-scaling.json
{
"AutoScalingGroupName": "ec2-workshop-asg",
"PolicyName": "automaticScaling",
"PolicyType": "TargetTrackingScaling",
"EstimatedInstanceWarmup": 300,
"TargetTrackingConfiguration": {
"PredefinedMetricSpecification": {
"PredefinedMetricType": "ASGAverageCPUUtilization"
},
"TargetValue": 75,
"DisableScaleIn": false
}
}
EoF
aws autoscaling put-scaling-policy --cli-input-json file://asg-automatic-scaling.json
Command should return policy ARN and target tracking alarms that have been created.
{
"PolicyARN": "arn:aws:autoscaling:ap-southeast-2:115751184547:scalingPolicy:04b7b7eb-6d65-40fb-946d-e5d2a1a55747:autoScalingGroupName/ec2-workshop-asg:policyName/automaticScaling",
"Alarms": [
{
"AlarmName": "TargetTracking-ec2-workshop-asg-AlarmHigh-6c60b9c6-b7e8-4fcf-9733-d9c390754b99",
"AlarmARN": "arn:aws:cloudwatch:ap-southeast-2:115751184547:alarm:TargetTracking-ec2-workshop-asg-AlarmHigh-6c60b9c6-b7e8-4fcf-9733-d9c390754b99"
},
{
"AlarmName": "TargetTracking-ec2-workshop-asg-AlarmLow-70cbbd68-5540-4293-a4c5-3ab2d8aa72bb",
"AlarmARN": "arn:aws:cloudwatch:ap-southeast-2:115751184547:alarm:TargetTracking-ec2-workshop-asg-AlarmLow-70cbbd68-5540-4293-a4c5-3ab2d8aa72bb"
}
]
}
Because predictive scaling starts launching instances only beginning of each hour, you have to wait for the end of the hour to see dynamic scaling in action alongside predictive scaling. To save time you can proceed with this next chapters.
Now you have predictive and dynamic scaling polices in place to ensure application responsiveness at the times of increased demand. You have successfully completed the first task you’ve been asked to do. Great work!