Back to Blog

Zoe Nageotte

Don't Strain Anything: VMTurbo and Stretched Clusters


As I organize my virtual environment, how do I maximize the availability of my resources?

This is a good question to ask yourself as you map your virtualized environment and quite frankly one that has multiple answers. Stretched clustering is one way that businesses can combine resources from separate geographic host locations and maintain continuous data availability in case of downtime or host failure. With ease of virtual machine migration across separate physical host sites, a stretched cluster enables high availability and downtime avoidance.

Cool. We have stretched clusters. So I’m all set then?

Well, no, not quite.

Let’s say you have two datacentres in separate locations but you have enabled stretch clustering so that the hosts act as a single pooled system. If a host or two goes down in one datacentre, they can failover onto the available hosts in the second datacentre. This is handy because critical applications can remain available as their virtual machines can be quickly and easily migrated within the virtual cluster, even though the hosts are located in different places.

However, in order for this scenario to successfully occur, the second datacentre must have enough capacity available to host these critical workloads after host failure occurs in the first datacentre.

Additionally, these pooled metrics can cause the virtual environment to become that much more complex, and maintaining a healthy state can be that much more challenging.

Agh! How do I ensure capacity while sustaining a healthy virtual environment?

Here is where VMTurbo comes in. VMTurbo’s demand-driven control platform immediately understands the workload resource demand and the constraints within a stretched cluster. It also actively makes the appropriate decisions to keep the environment at its healthiest state where application performance is assured while utilization is maximized. VMTurbo can identify when additional resources need to be added—like adding a new host—so that you can continue to maintain this business policy.

Furthermore, by recommending or taking actions, VMTurbo will assure that adequate capacity is maintained at both datacentre sites.

For example, if VMTurbo sees that the memory of all of your hosts across your environment is at 57%, it will notify you to add more hardware since you are violating your business policies.

Wait, 57% memory isn’t bad though!

Maybe not normally, but if a stretched cluster is over 50% in overall capacity and one datacentre goes down, then not all virtual machines will be able to come back on in the second datacenter and data could be lost. In the case of using stretched clusters as a failover plan, you need to ensure that you have sufficient space for all VMs.

So, how exactly does VMTurbo assure performance in stretched clusters?

VMTurbo’s control platform supports stretched cluster policies by managing the complexity of pooled resources through actively distributing workloads to minimize risk and maintain a desired state within the virtual environment. With application demand being aligned to the appropriate resource supply, workload satisfaction and performance will always be assured.

Regardless of whether you would automate these actions or execute them manually, the platform will still recommend the required next steps to assure the health of your environment.

Want to see how VMTurbo’s Operations Manager can help to manage your environment? Download our free 30-day trial here.