Business Scenarios

It is suitable for scenarios where the visits or traffic of user's business fluctuate obviously and increase greatly at peak hours and drops during other periods. Computing resources are idle in off-peak hours, resulting in cost waste.

In this case, you can use AS to create a low-cost service cluster and automatically scale up or down computing resources in the cluster based on service traffic to reduce cost.

Create an AS group

Log in to the AS Console;
On the top of the page, select a Region;
Click to create an AS group;
Set the minimum number of instances to 0;
Select the owning VPC and subnet of the service cluster;
Bind the load balancer used by the service module based on business requirements;
Select later created instances for the instance removal policy;
Select the release recycling mode for pay-per-use instances;
Click Next;
Select Create Now for AS configuration;
Select ECS specifications and system disk and data disk capacities based on business requirements;
Select a private image as the image type, and select the image corresponding to the service module;
Configure EIP and passwords based on business requirements
Click Create Now;

Create an ECS instance with annual or monthly package using a custom image

Create a certain number of ECS instances with annual or monthly package based on service requirements to ensure the normal operation of the service cluster in off-peak hours;
Log in to the AS Console;
In the pop-up window, select the ECS instances created in step 1, and click OK;
In the list of AS instances, select the added ECS instances, and click Enable Instance Protection;
Set AS policies (capacity expansion);
Click Modify on the details page of AS group;
On the AS group modification page, modify the minimum and maximum instances of the AS group based on business requirements;
At the bottom of the details page of the AS group, click Create Policy;
Select the alarm strategy for the policy type.
Select Create Now for alarm rules;
Set the trigger condition to the average CPU usage greater than or equal to 50%;
Set the monitoring period to 5 minutes;
Set the number of consecutive occurrences to 3;
Set the cooldown time to 300s;
Set the execution action to Add 5 Instances;

Set AS policies (capacity reduction)

At the bottom of the details page of the AS group, click Create Policy;
Select the alarm strategy for the policy type;
Select Create Now for alarm rules;
Set the trigger condition to the average CPU usage smaller than or equal to 30%;
Set the monitoring period to 5 minutes;
Set the number of consecutive occurrences to 3;
Set the cooldown time to 300s;
Set the execution action to Reduce 5 Instances;

Action Results

The ECS instances with annual or monthly package in the AS group are set to the protected state. A protected ECS will not be automatically removed from the AS group by the AS policy, ensuring the normal operation of the service cluster in off-peak hours. During peak hours, the AS group automatically increases or decreases the number of ECS instances based on the alarm rules set by the AS policy, reducing the idle time of computing resources and cost waste.

Auto Scaling

Use AS to Create Low-cost Service Clusters