Introduction to Cloud Bursting and Action plans in Abiquo

According to Gartner, “Cloud bursting is the use of an alternative set of public or private cloud-based services as a way to augment and handle peaks in IT system requirements at startup or during runtime. Cloud bursting can span between on-premises IT systems and services and the cloud, across multiple cloud providers or across multiple resource pools of a single provider. It can also be enabled across multiple internal data centers, across multiple external data centers, or between internal and external data centers.”

But what does that really mean for your business? Cloud bursting helps you add resilience to avoid service interruption and, consequently, helps to reduce service outage costs, protect revenue and ensure customer retention by meeting your SLAs.

In this post, we’ll take you through the steps to set up cloud bursting in Abiquo 4.4.

In previous versions of Abiquo, users could create alerts and alarms in order to trigger automated action sequences (“Action Plans”) , including deploying VMs or initiating scaling actions in the same VDC when they reached configured thresholds of guest utilization or application performance.

The new Control panel together with new self-service VPN features mean that users can use Action plans for cloud bursting to deploy new VMs or trigger scaling actions in any Abiquo-managed cloud, thus enabling cloud users to “burst” workloads into private data centers and public clouds and ensure they are connected to the necessary networks, firewalls and loadbalancers.

How it works

Using Abiquo 4.4 users can use Action plans for cloud bursting and, so, set up alarms that will make possible to deploy in private or public datacenters when VMs reach a peak in demand.

Users can create action plans that act on multiple elements in throughout the cloud. This means that you can implement cloud bursting driven by application monitoring.

To set up cloud bursting, follow these steps. We’ve included links to our documentation so you can explore each step in more detail:

  1. Ensure your application is running in an Abiquo Virtual Datacenter (VDC), and set up scaling groups which will later be activated by your action plans.
  2. Set up monitoring for all the resources that will drive your cloud bursting activity. This could be as simple as CPU/RAM for one VM, or a combination of any monitored metrics across your guests (including any metrics pushed into Abiquo from your application) and infrastructure monitoring.
  3. For each metric, set up two alarms – one for when your desired threshold is exceeded, and one for when it’s below a “normal” level. The first will be used to trigger the alerts for bursting and the second to remove the added resources.
  4. In the new Control page, set up alerts – groups of alarms that all need to be active to trigger an action – by grouping the “threshold exceeded” alarms together, and the “normal” alarms together. These alerts will trigger the action plans.  
  5. Now choose the target for the scaling actions. This could be a scaling group in any VDC for the tenant, or a single VM which you might leave powered off – whatever suits your needs. We’ll use a scaling group for the rest of this example.
  6. In the Control page also, set up Action Plans, one for bursting, and one to remove it. In the Action Plan, give it a name and description and start adding actions. The categories of action include general actions (email), VM actions (e.g. power on) and scaling group actions (e.g. scale out).

You can, of course, get more sophisticated as you need:

  • Test your action plans with the Execute button
  • Set up schedules for when scaling groups should respond to action plans
  • Drive action plans from scheduled as well as (or instead of) monitoring data
  • Add other actions, such as scale up/down, adding resources to individual VMs, restarting VMs and more.

Contact us for a demo if this is something you could use!

7. For each action, select the entity to act on, as shown in the screenshot below where we’ve selected a scaling group in another VDC, then choose the action. To add VMs to a scaling group, simply select “Scale out”. To remove VMs, select “Scale in”. The number of VMs added, and the pause between new VMs being created, is set in the scaling group itself as before.  
8. Then add any other actions. You may want to send an email, to reboot a VM with CPU at 100% for a specified time, add CPU to a VM. You can do it all in one plan, or use multiple plans.

9. Finally, specify the alerts you created earlier to trigger the plans. You can also add a schedule to trigger the plans at a defined date/time, if you can predict a spike in resource needs.


Networking and cloud bursting

Abiquo’s scaling groups already add new VMs to the same networks as the master VM in the group, and also apply the same firewall rules and add the VM to any configured load balancers. If you want to burst in the same VDC, that’s all you need. To burst between VDCs in datacenters that are connected already, you could use an External network to connect them together.  This would work for your own on-premise infrastructure, or for private/public VDCs where you already have a service such as Amazon Direct Connect which would present as an External Network in Abiquo.

Where you don’t have such connectivity, you can use the new VPN feature in Abiquo 4.4.  The lets you create VDC-to-VDC VPNs through the Abiquo GUI. This is supported in AWS and VMWare NSX, with Azure and Neutron coming soon. By creating a VPN you can route traffic through a firewall and loadbalancer onto your set of resources across clouds.


With cloud bursting, Abiquo users now can benefit from an easier and faster way to add resources from multiple cloud providers when higher capacity is required and so reach a more complete multi-cloud strategy.