Autoscaling Groups of Instances

Managed instance groups offer autoscaling capabilities that allow you to automatically add or delete instances from a managed instance group based on increases or decreases in load. Autoscaling helps your applications gracefully handle increases in traffic and reduces cost when the need for resources is lower. You just define the autoscaling policy and the autoscaler performs automatic scaling based on the measured load.


Autoscaling works by adding more instances to your instance group when there is more load (upscaling), and deleting instances when the need for instances is lowered (downscaling).


Autoscaling uses the following fundamental concepts and services.


Managed instance groups

Autoscaling is a feature of managed instance groups. A managed instance group is a pool of homogeneous instances, created from a common instance template. An autoscaler adds or deletes instances from a managed instance group. Although Compute Engine has both managed and unmanaged instance groups, only managed instance groups can be used with autoscaler.


To understand the difference between a managed instance group and unmanaged instance group, see the Instance Groups documentation

Autoscaling policy and target utilisation

To create an autoscaler, you must specify the autoscaling policy and a target utilisation level that the autoscaler uses to determine when to scale the group. You can choose to scale using the following policies:


  • Average CPU utilisation
  • HTTP load balancing serving capacity, which can be based on either utilisation or requests per second.
  • Stackdriver Monitoring metrics


The autoscaler will collect information based on the policy, compare it to your desired target utilisation, and determine if it needs to perform scaling.


The target utilisation level is the level at which you want to maintain your virtual machine instances. For example, if you scale based on CPU utilisation, you can set your target utilisation level at 75% and the autoscaler will maintain the CPU utilisation of the specified group of instances at or close to 75%. The utilisation level for each metric is interpreted differently based on the autoscaling policy.



Autoscaling only works with managed instance groups. Unmanaged instance groups are not supported.


Do not use Compute Engine autoscaling with managed instance groups that are owned by Kubernetes Engine. For Kubernetes Engine groups, use Cluster Autoscaling instead.


An autoscaler can make scaling decisions based on multiple metrics, but it can handle only one policy per metric type except in the case of Stackdriver monitoring metrics; an autoscaler can handle up to five policies based on Stackdriver monitoring metrics. The autoscaler calculates the recommended number of virtual machines for each policy and then scale based on the policy that provides the largest number of virtual machines in the group.

Want To Know More​

Need further information or require a quotation?

All calls are recorded for security, training and quality purposes

Our lines are open Monday to Friday from 9am to 5.30pm. Dialling an 0330 number costs the same to dial as a call to a geographic (local) number. They cost the same to call from a landline or mobile and are included in mobile call packages.

You are calling our Network Operations Centre based in London, United Kingdom.

Just so you know, we are not able to accept telesales or telemarketing calls and can't be transferred.

Working proudly with skilled teams of people knowing we push the boundaries staying ahead of the curve producing high performance results.