Manage scaling

Contents 

Overview

On the Manage page, the Scale & Redundancy tab lets you configure manual or automatic scaling of your deployment. Blueprints that include steps for scaling resources up or down (such as adding or removing servers) are considered self-contained blueprints. 

If scaling is not configured when you deploy the blueprint, you can configure scaling after deployment using the Scale & Redundancy tab under Manage. You can also edit or modify existing scaling processes associated with your blueprint deployment. Set up scaling options to scale the blueprint depending on your system behavior. You can perform manual or auto scaling depending on your requirements. Depending on the scaling option you select and configure, the scaling process details appear in the Scale & Redundancy tab under Manage.

Manual scaling

The Blueprint Designer provides a convenient Scale Up/Scale Down function that lets you manually scale your server and application resources up and down as needed.

Smart Tools provides the following manual scaling options:

  • Scale Up: Use this option to run a process that adds resources (servers and applications) to a deployment launched through an initial process. You can also use Scale Up to add resources (such as applications) to pre-existing servers provisioned independently.
  • Scale Down: Use this option to remove resources from a deployment (stop, remove, terminate servers and applications). The Scale Down option runs a preconfigured shutdown process containing steps for removing resources.

Auto scaling

The Blueprint Designer provides a powerful way to automatically scale your deployment based on your requirements. You can configure rules and preferences that govern when the deployment should be scaled up or down and how many servers must be added or removed. 

View scaling processes

  1. From the Deployments page, move the pointer over the deployment you want to view scaling processes and click Details.
  2. Click the Scale & Redundancy tab. A list of scaling processes is displayed. You can view information such as scaling process name, status, and run schedule. You can add scale up and scale down processes from this page. 
  3. Click the name of a scaling process to view a summary on the scaling process deployment. The scaling process run information, VMs and components involved, timestamp of when the scaling process is run, server names, and IP addresses are displayed.
  4. Click Get Deployment Report to view deployment details of a scaling process such as status, step details, step outputs and results. 
  5. To go back to scaling processes list, click Runs.

Configure manual scaling

  1. From the Deployments page, move the pointer over the deployment you want to configure scaling and click Details.
  2. Click the Scale & Redundancy tab.
  3. Move the pointer below the Scale & Redundancy tab and click Configure.
    The deployed blueprint opens in the edit mode. The Scale & Redundancy and Tear Down tabs are active and other tabs are inactive in edit mode.
  4. Select This is a self contained process that can be scaled up / down or decommissioned
  5. Select Manual Scaling.
  6. Click Save to save the manual scaling updates made to the blueprint.
    The blueprint instance is updated and the Scale & Redundancy tab appears.
  7. To scale up your deployment, click Scale Up and select one of the following scaling types: 
    • Scale Up by Running Again: Re-runs the process to provision and configure new machines. To configure this scaling type, perform the following actions:
      1. Select Scale Up by Running Again and click Next.
      2. Review the server launch steps that will be used to scale this process and click Next.
      3. Review blueprint parameters (if applicable) and click Finish.
    • Scale Up by adding existing servers to this process: Uses existing machines that were not part of the blueprint deployment. To configure this scaling type, perform the following actions:
      1. Select Scale Up by adding existing servers to this process and click Next.
      2. Select a server or multiple servers that will be used to scale this process and click Next. 
      3. Review blueprint parameters (if applicable) and click Finish.
  8. To scale down, click Scale Down. In the Scale Down Server dialog, select the servers to scale down, review blueprint parameters, and then click Finish. 

The configured scale up and scale down processes appears in the list of processes. Click Refresh if you are unable to see a configured process. To view information such as scaling process name, status, and run schedule, click the name of a scaling process.

Configure auto scaling

  1. Move the pointer below the Scale & Redundancy tab and click Configure.
    The deployed blueprint opens in the edit mode. The Scale & Redundancy and Tear Down tabs are active and other tabs are inactive in the edit mode.
  2. Select This is a self contained process that can be scaled up / down or decommissioned
  3. Select Auto Scaling (Rule Based).
  4. Select scaling rule conditions (All or Any) from the If box. The default option is Any.
  5. Click Add Rule to add a scaling rule. The Add Rule dialog appears.
  6. Select the basis for scaling from the For box and click Next.
    The default condition is Avg of All Machines in Blueprint.
  7. Select a metric from the Metric For Scaling box that will scale this process up or down.
    This can either be an app, system or custom metric. For more information on metrics, see the Monitor servers and applications topic.
  8. Set up conditions to scale up the process.
    When this criteria is met, the process will be scaled up.
  9. Set up conditions to scale down the process.
    When this criteria is met, the process will be scaled down. Any steps defined in the shutdown process will be run in order. Machines launched by a previous scale up operation will be terminated.
  10. Click Finish to save and add the rule to the auto scaling process.
  11. (Optional) Select Do not initiate Scale Up when a previous Scale Up is pending to direct Smart Tools to wait for the completion of an already initiated scale up process, before the next scale up process is run.
    This can be useful when a scale up process takes a longer time than the definition of the thresholds.
  12. Enter the minimum time to wait (in minutes) before launching the next scale up operation.
    Setting this value will ensure that the scale up operation waits for this time period to elapse even if the thresholds indicate a need to scale up.
  13. Enter the minimum time to wait (in minutes) before launching the next scale down operation.
    Setting this value will ensure that the scale up operation waits for this time period to elapse even if the thresholds indicate a need to scale down.
  14. Enter the minimum number of servers to launch as part of the scale up operation.
    This establishes a minimum number of machines below which the stack will not be scaled down in order to prevent all the machines in the stack from being deleted.
  15. Enter the maximum number of servers that will be present in the process, in order to prevent run away scaling.
  16. To scale down machines, a tear down step is also required. Click the Tear Down tab, drag and drop the Terminate Machine step from the Cloud Steps to the tear down process steps.
  17. Click Save to save the auto scaling updates made to the blueprint. The Scale & Redundancy tab under Manage appears with the updated scaling information.
  18. To pause auto scaling process, click Pause Auto Scaling. To resume auto scaling, click Start Auto Scaling. To Edit scaling policy, click Edit.

You can view the auto scaling configuration details such overview, scaling policy that includes scaling rules and preferences, and  logs by clicking the respective tabs.

0 Comments