LATEST VERSION: 1.10 - CHANGELOG
Pivotal Cloud Foundry v1.7

Configuring Your App Autoscaling Instance

Page last updated:

The App Autoscaling service scales bound applications in response to load.

An instance of the App Autoscaling service examines the CPU usage of an application bound to it every few minutes. In response to load changes, the service scales your app up and down according to the thresholds, minimums, and maximums that you provide.

Follow the steps below to configure your App Autoscaling service instance.

  1. Log in to the Apps Manager: Logging into the Apps Manager

  2. Select a space containing an App Autoscaling service instance from the org dashboard or from the left navigation bar.

  3. In the Services section of the space dashboard, under your App Autoscaling service instance name, click Manage.

    Note: You must specifically have the role of Space Developer to access the Manage link for the app autoscaling service. Space Managers, Space Auditors, and all Org roles do not have the permission to make changes to App Autoscaling. See Managing User Accounts and Permissions Using the Apps Manager for help managing user roles.

    0 manage instance

  4. By default, new autoscaling instances are paused. If paused, click Turn On to enable monitoring of your application and start the autoscaling process.

    Paused Instance

  5. Click the wrench icon on your Autoscaling dashboard.

    Configure Autoscaling

  6. Change the configuration settings and click Save. See the Configuration Options section of this topic for information about the configuration settings.

    Edit Configuration

  7. Examine the App Autoscaling service instance dashboard to confirm your changes.

    View Changes

Configuration Options

You can set the absolute maximum and minimum number of instances for your app, as well as the CPU thresholds for an app that trigger the autoscaling service.

Instance Counts

The Instances values specify the absolute minimum and maximum number of instances autoscaling can set for an application.

  • Min: Default value: 2. The minimum number of instances to which autoscaling can scale your app. Autoscaling never scales your application below this number of instances.

  • Max: Default value: 5. The maximum number of instances to which autoscaling can scale your app. Autoscaling never scales your application above this number of instances.

Note: Min and Max cannot be set to less than 1 or greater than 20. Min must be less than or equal to Max.

CPU Thresholds

The CPU thresholds values specify the upper and lower limits of CPU utilization that trigger the autoscaling service.

The autoscaling service calculates CPU utilization as a moving average across the CPUs of all currently running instances of an application.

  • Low: Default value: 20. When the autoscaling service instance detects CPU utilization below this threshold, it reduces the number of instances of the app by one.

  • High: Default value: 80. When the autoscaling service instance detects CPU utilization above below this threshold, it increases the number of instances of the app by one.

Manual Scaling

If you manually scale an application bound to an autoscaling service instance, the autoscaling service stops monitoring and autoscaling your application.

To re-enable monitoring and scaling, click Turn On on the App Autoscaling service instance dashboard.

Paused Instance

Create a pull request or raise an issue on the source for this page in GitHub