Scaling based on CPU utilization


The simplest form of autoscaling is to scale a managed instance group (MIG) based on the CPU utilization of its instances.

You can also autoscale a MIG based on the load balancing serving capacity, Monitoring metrics, or schedules.

Before you begin

  • Review the autoscaler limitations.
  • Read about autoscaler fundamentals.
  • If you haven't already, set up authentication. Authentication is the process by which your identity is verified for access to Google Cloud services and APIs. To run code or samples from a local development environment, you can authenticate to Compute Engine as follows.

    Select the tab for how you plan to use the samples on this page:

    Console

    When you use the Google Cloud console to access Google Cloud services and APIs, you don't need to set up authentication.

    gcloud

    1. Install the Google Cloud CLI, then initialize it by running the following command:

      gcloud init
    2. Set a default region and zone.

    REST

    To use the REST API samples on this page in a local development environment, you use the credentials you provide to the gcloud CLI.

      Install the Google Cloud CLI, then initialize it by running the following command:

      gcloud init

Scaling based on CPU utilization

You can autoscale based on the average CPU utilization of a managed instance group (MIG). Using this policy tells the autoscaler to collect the CPU utilization of the instances in the group and determine whether it needs to scale. You set the target CPU utilization the autoscaler should maintain and the autoscaler works to maintain that level.

The autoscaler treats the target CPU utilization level as a fraction of the average use of all vCPUs over time in the instance group. If the average utilization of your total vCPUs exceeds the target utilization, the autoscaler adds more VM instances. If the average utilization of your total vCPUs is less than the target utilization, the autoscaler removes instances. For example, setting a 0.75 target utilization tells the autoscaler to maintain an average utilization of 75% among all vCPUs in the instance group.

You can also scale based on forecasted CPU utilization. For more information, and to see if this is suitable for your workload, see Scaling based on predictions.

Enable autoscaling based on CPU utilization

Console

  1. In the console, go to the Instance groups page.

    Go to Instance groups

  2. If you have an instance group, select it and click Edit. If you don't have an instance group, click Create instance group.

  3. If no autoscaling configuration exists, under Autoscaling, click Configure autoscaling.

  4. Under Autoscaling mode, select On: add and remove instances to the group to enable autoscaling.

  5. Specify the minimum and maximum numbers of instances that you want the autoscaler to create in this group.

  6. In the Autoscaling metrics section, if an existing CPU utilization metric does not yet exist, add one:

    1. Click Add metric.
    2. Under Metric type, select CPU utilization.
    3. Enter the Target CPU utilization that you want. This value is treated as a percentage. For example, for 75% CPU utilization, enter 75.
    4. Under Predictive autoscaling, select Off. To learn more about predictive autoscaling, and whether it is suitable for your workload, see Scaling based on predictions.
    5. Click Done.
  7. You can use the Initialization period to tell the autoscaler how long it takes for your application to initialize. Specifying an accurate initialization period improves autoscaler decisions. For example, when scaling out, the autoscaler ignores data from VMs that are still initializing because those VMs might not yet represent normal usage of your application. The default initialization period is 60 seconds.

  8. Click Save.

gcloud

Use the set-autoscaling sub-command to enable autoscaling for a managed instance group. For example, the following command creates an autoscaler that has a target CPU utilization of 60%. Along with the --target-cpu-utilization parameter, the --max-num-replicas parameter is also required when creating an autoscaler:

gcloud compute instance-groups managed set-autoscaling example-managed-instance-group \
    --max-num-replicas 20 \
    --target-cpu-utilization 0.60 \
    --cool-down-period 90

You can use the --cool-down-period flag to set the initialization period, which tells the autoscaler how long it takes for your application to initialize. Specifying an accurate initialization period improves autoscaler decisions. For example, when scaling out, the autoscaler ignores data from VMs that are still initializing because those VMs might not yet represent normal usage of your application. The default initialization period is 60 seconds.

Optionally, you can enable predictive autoscaling to scale out ahead of predicted load. To learn whether predictive autoscaling is suitable for your workload, see Scaling based on predictions.

You can verify that autoscaling is successfully enabled by using the instance-groups managed describe sub-command, which describes the corresponding managed instance group and provides information about any autoscaling features for that instance group:

gcloud compute instance-groups managed describe example-managed-instance-group

For a list of available gcloud commands and flags, see the gcloud reference.

REST

To create an autoscaler, use the autoscalers.insert method for a zonal MIG or the regionAutoscalers.insert method for a regional MIG.

The following example creates an autoscaler for a zonal MIG:

POST https://compute.googleapis.com/compute/v1/projects/PROJECT_ID/zones/ZONE/autoscalers/

Your request body must contain the name, target, and autoscalingPolicy fields. autoscalingPolicy must define cpuUtilization and maxNumReplicas.

You can use the coolDownPeriodSec field to set the initialization period, which tells the autoscaler how long it takes for your application to initialize. Specifying an accurate initialization period improves autoscaler decisions. For example, when scaling out, the autoscaler ignores data from VMs that are still initializing because those VMs might not yet represent normal usage of your application. The default initialization period is 60 seconds.

Optionally, you can enable predictive autoscaling to scale out ahead of predicted load. To learn whether predictive autoscaling is suitable for your workload, see Scaling based on predictions.

{
  "name": "example-autoscaler",
  "target": "https://www.googleapis.com/compute/v1/projects/myproject/zones/us-central1-f/instanceGroupManagers/example-managed-instance-group",
  "autoscalingPolicy": {
    "maxNumReplicas": 10,
    "cpuUtilization": {
      "utilizationTarget": 0.6
    },
    "coolDownPeriodSec": 90
  }
}

For more information about enabling autoscaling based on CPU utilization, complete the tutorial, Using autoscaling for highly scalable apps.

How autoscaler handles heavy CPU utilization

During periods of heavy CPU utilization, if utilization is close to 100%, the autoscaler estimates that the group might already be heavily overloaded. In these cases, the autoscaler increases the number of virtual machines by 50% at most.

What's next