Enabling auto scaling health checks for AWS VPCs

This topic describes how cloud administrators can create auto scaling health checks (using the BMC Cloud Lifecycle Management Service Monitoring feature) to horizontally scale out or scale in a Virtual Private Cloud (VPC). The topic includes the following sections:

• • Overview of auto scaling support for AWS
  • To enable auto scaling health checks
  • To create a monitoring policy
  • Example rule

Overview of auto scaling support for AWS

Implementing auto scaling enables you to scale your Amazon Web Services (AWS) capacity up or down automatically, according to conditions you define using the BMC Cloud Lifecycle Management Service Monitoring feature. With auto scaling, you can ensure that the resources of the AWS instances in use increase seamlessly during demand spikes to maintain performance, and decrease automatically during demand lulls to minimize costs. Auto scaling is useful for applications that experience variability in usage. 

• Auto scaling is available for the following service offering instance (SOI) types:
  • Gold level (service blueprints that contain at least one Network Path and possibly Load Balancing objects)
  • Silver level (service blueprints that contain Load Balancing objects)
  • Bronze level (any basic Service Blueprint). 
• Administrators can adjust the following options for auto scaling:
  • Poll time
  • CPU and memory size
  • Server count (add/remove server)

To enable auto scaling health checks

To implement auto scaling for AWS, you must enable monitoring, select a monitory level, and enable a policy in the service blueprint deployment definition.

1. In the Service Designer workspace, select a service blueprint from the Blueprint Library.
   The service blueprint appears in view-only mode.
2. Click Check Out.
   A local copy of the service blueprint is created and is available to edit.
3. Click Definition > Properties.
   The Definition Details dialog box opens.

4. On the Properties tab, enter information for the following fields:

   Field	Action
   Enable Monitoring	Select the Enable Monitoring check box to enable performance monitoring of the resources provisioned by the service blueprint.
   Monitoring Level	Select a Monitoring Level.
   Enable Monitoring Policy	Select the Enable Monitoring Policy check box to allow users to create monitoring rules in the My Services console.

   After monitoring is enabled, the My Services console reflects the AWS CloudWatch metrics in the service health charts. 

5. Select File > Save and Check In.
6. Click OK.
   The blueprint is removed from the My Checked Out Blueprints list, and is added to the Blueprint Library.

To create a monitoring policy

You can create a policy to take remedial actions based on how the monitored metrics are interpreted.

1. On the Services link bar of the My Services tab, select a row in the Services table.
   If the service has service monitoring policies enabled, the Manage Policies icon  appears in the toolbar.
2. Click the Manage Policies icon.
   The Policies display opens, showing the existing policies for the selected service. By default, the Policies display is empty.
3. In the Policies display, click New Rule.
   The Add a new rule dialog box opens.
4. In the Rule Name field, enter a name for the rule.
5. In the When section, define one or more conditions that must be met to trigger an action:
   
   1. Select a type of metric.
   2. Select a comparative symbol (>, >=, <, <=, or =).
   3. Enter a number for the threshold that triggers an action.
   4. Enter the number of minutes the threshold must be reached to trigger an action.
   5. If you want to add another condition, click Add Clause, specify whether the new condition is additional (AND) or optional (OR), and then repeat steps a through d.
6. In the Do this section, select Perform an action.

7. In the drop-down list, select the action you want to perform.
   The following table lists the actions that apply to AWS.

    
8. Click OK.
   The new rule appears at the bottom of the list of rules.
9. If necessary, change the precedence of the rule:
   1. Drag and drop the rule to a new place in the list of rules.
   2. Click Apply Precedence.
10. When you are ready for the rule to go into effect, click Activate.

Example rule

For example, you could create a policy that states when CPU activity is higher than 80% for more than 15 minutes, add a server. This action would automatically scale out the VPC,  

Related topics

Managing-service-monitoring