¶ About AutoScale VM Groups
¶ What is AutoScaling?
AutoScaling VM is a feature that allows cloud-based infrastructure to automatically adjust the number of virtual machines in response to changes in demand or workload. When demand increases, additional virtual machines are automatically provisioned to handle the load, ensuring optimal performance and availability. Conversely, when demand decreases, excess virtual machines are automatically terminated to save costs. This dynamic scaling capability helps maintain consistent performance while minimizing resource waste and manual intervention.
¶ Requirements
Consider the following requirements before creating an AutoScale VM Group:
- Before configuring AutoScale, make sure the required template is prepared. Also, ensure that when an instance is deployed using a template, the application is up and running upon instance launch.
- Create an isolated network using a network offering that supports instance AutoScaling. Acquire a new IP address to be used as the source NAT of the network, and create a load balancer rule without associating it with any instance.
By default, network offerings supporting load balancers have AutoScaling enabled. To disable it, create a network offering without AutoScaling support.
¶ Creating AutoScale VM Groups
¶ Adding new AutoScale VM Groups
To add a new AutoScale VM Group, follow these steps:
1. Navigate to Compute → AutoScale VM Groups.
2. Here, you can view all previously created AutoScale Groups. To add a new group, click on New AutoScale VM Group in the top row:
3. First, choose a deployment zone for your instance. The options you can select from are WSB1, FRA1 and FRA2.
4. Next, select an Template from the list. You can choose from a wide variety of operating systems, applications, or custom images.
5. Now, choose a Compute offering according to your needs. The hardware offerings vary in pricing, CPU cores, and RAM memory. You can find more information about that in our service offerings.
6. Optionally, select an additional Data disk.
7. Choose at least one Network for VMs in the AutoScale VM group. The default network must be an isolated network or VPC tier which supports VM AutoScaling and has load balancing rules.
8. Select a Load balancer rule for your AutoScale VM group. Please note that load balancing must be configured for your Public IP address.
9. The next two steps involve configuring the ScaleUp and ScaleDown policies of the VM group. If all the conditions are met, the AutoScale VM Group will either scale up or down accordingly. For more information, refer to section AutoScale Policies.
10. You can also configure Advanced settings if desired, such as SSH Key Pairs, Affinity Groups, and User Data.
11. Lastly, you have to configure some minor Details:
- Select a Name for your AutoScale VM Group.
- Define the Expunge VM grace period, which sets the time between scaling down and expunging an instance, with the default being 120 seconds.
- Specify the number of Max members in the VM Group, which corresponds to the upper limit of instances that can be assigned to the load balancer rule.
- Specify the number of Min members in the VM Group, serving as a lower limit to ensure that the load balancer has at least a specific number of active instances.
- Define a Polling interval at which the performance counters are evaluated before scaling up or down is initiated, with the default set to 30 seconds.
12. Click on Create to save and add your new AutoScale VM Group.
¶ AutoScale Policies
As introduced in the previous chapter, a VM Group must have AutoScale Policies, specifically at least one ScaleUp and one ScaleDown policy. These policies include the following parameters:
- Name: a name for the AutoScale policy.
- Duration: time frame in which the conditions must be true before action is taken.
- Quiet time: cool down period during which the policy should not be evaluated after the action has been taken.
- Condition: an AutoScale policy must include at least one condition.
On the other hand, each condition has the following parameters:
- Counter: the resources for which the policy will be evaluated, including the following ones:
- VM CPU - average percentage
- VM RAM - average percentage
- Public Network - mbps received per Instance
- Public Network - mbps transmit per Instance
- Load Balancer - average connections per Instance
- Operator: the relation operator to be used with the threshold, such as greater than, less than, etc.
- Threshold: a value for which the counter will be evaluated with the operator selected.
¶ Managing AutoScale VM Groups
Almost every configuration for the AutoScale VM Groups explained so far can still be edited and managed in the detail view of the instance group. If you plan on adjusting or maintaining the AutoScale VM Groups, we strongly recommend disabling the AutoScale Group beforehand. In the disabled status, no scale up or scale down action is performed, allowing for seamless performance of any maintenance activity. To disable an AutoScale VM Group, follow these easy steps:
1. Navigate to Compute → AutoScale VM Groups.
2. Select the VM Group you wish to manage from the list view.
3. Click on Disable AutoScale VM Group in the top right action toolbar:
After your maintenance activity is completed, you can activate the VM Group in the same way by clicking on Enable AutoScale VM Group this time. The adjustments will take effect and the VM Group will be ready to go again.
¶ Updating AutoScale VM Groups
You can modify various AutoScale VM Group settings after its creation. To do so, follow these steps:
1. Navigate to Compute → AutoScale VM Groups.
2. Select the VM Group you wish to update from the list view.
3. Click on Update AutoScale VM group in the top right action toolbar:
4. Update Name, Max members, Min members, and Polling interval.
5. Click on OK to save the changes.
¶ Updating AutoScale VM Profile
You can also update certain AutoScale VM Profile settings. To accomplish that, follow these steps:
1. Navigate to Compute → AutoScale VM Groups.
2. Select the VM Group you wish to update from the list view.
3. Click on AutoScale VM Profile in the right management block.
4. Clicking on the Edit AutoScale VM Profile button opens a window where you can modify the Expunge VM grace period, Template, and Compute offering of your AutoScale VM Group:
5. Clicking on the Reset Userdata on AutoScale VM Group button right next to it provides you with the option to update and change the User data of your VM Group:
6. Lastly, you can update specific Parameter settings for your AutoScale VM Group, such as the following ones:
- affinitygroupids: The UUID of the affinity groups, separated by a single comma character (,).
- diskofferingid: The UUID of the data disk.
- disksize: The size of data disk. This is valid only if the disk offering is dynamic.
- keypairs: The name of the SSH Key pairs, separated by a single comma character (,).
- networkids: The UUID of the Instance networks, separated by a single comma character (,).
- overridediskofferingid: The UUID of override disk offering for ROOT disk.
- rootdisksize: The size of the ROOT disk. This overrides the size of the Instance Template.
- securitygroupids: The UUID of security groups, separated by a single comma character (,).
¶ Updating AutoScale Policies
You can also add, update, and remove AutoScale Policies. To do so, follow these steps:
1. Navigate to Compute → AutoScale VM Groups.
2. Select the VM Group you wish to update from the list view.
3. Depending on which policy you want to edit, click either on ScaleUp policy or Scale Down policy in the right management block.
4. To add a new policy, click on Add policy and specify Name, Duration, Quiet time, Counter, Operator, and Threshold as explained in the AutoScale policies above:
5. To update an existing policy, Select the ScaleUp/ScaleDown policy from the drop down menu, click on the Edit button, modify the values of the parameters or conditions, and save by clicking the Apply button:
6. Finally, you can remove either specific conditions or entire policies by clicking on the Remove policy button:
¶ Deleting AutoScale VM Groups
If AutoScale VM Groups are no longer needed and you want to reclaim space, you can easily remove them by following these steps:
1. Navigate to Compute → AutoScale VM Groups.
2. Select the VM Group you wish to remove from the list view.
3. Click on Delete AutoScale VM Group in the top right action toolbar:
4. Optionally, you can toggle the Clean up button if you want to force-delete the VM Group. Note that doing so will directly expunge the resource with no possibility of recovery.
AutoScale VM Groups can only be removed if there are no instances in the group.