Disaster Recovery on Demand?

Over the past year, I’ve had a lot of conversations about disaster recovery solutions. There is a common theme about replicating on-premise workloads to the public cloud for the sole purpose of disaster recovery.

Many enterprise solutions provide this capability, including VMware Cloud on AWS, Google Cloud VMware Solution by CloudSimple, Microsoft Azure Stack, and Nutanix Clusters. We could go into each product in detail, but let’s table that discussion for now. But let’s also agree that the premise is the same for each, to allow customers to migrate workloads to and from private and public infrastructure seamlessly.

As perfect as this sounds, one drawback outweighs the benefits of building this type of hybrid cloud. Can you guess? If you’re thinking ‘cost,’ you’re correct!

The cost associated with running workloads in the public cloud on bare-metal is enough reason to steer away from the solutions mentioned above. Instead, we decide on an active-backup approach, both on-premise. Or, we choose an active-active method where both data centers are on-premise, and our active workloads span across both so you can failover to one or the other in the case a disastrous event. Active-active allows you to take advantage of your entire infrastructure purchase, rather than having half of it idle waiting for a devastating event.

There is nothing wrong with finding solutions to help minimize costs; it is universal. That is what we are all asked to do. When I managed a large data center, it was critical to have a disaster recovery solution that quickly allowed us to failover with limited application interruption. Simultaneously, we did not have an unlimited budget, so we had to do more with less. Therefore, often sacrificing to balance functionality and cost. And with that also came complexity!

All of this got me thinking. Is there a way to leverage the public cloud, but only when you need it—something like, Disaster Recovery on Demand?

I immediately started thinking about the capabilities of Nutanix. Natively, AOS can replicate snapshots to AWS S3 Buckets or Azure S3 Blob storage. But, the ability to restore these snapshots in the public cloud is not possible. But, what about using a third-party backup solution. What about HYCU?

HYCU is purpose-built for Nutanix and supports both AHV and VMware ESXi. There is more we can dive into regarding HYCU, and we will in a future post, but for now, I want to dive into what HYCU can do with backing up VMs running on AHV in conjunction with their support for AWS S3 Buckets. Why AWS when HYCU also supports GCP and Azure? In short, it is what I know, and Nutanix Clusters is supported in AWS only, for now!

When we combine HYCU with Nutanix AHV, AWS S3, and Nutanix Clusters, you get what I called Disaster Recovery on Demand.

If you deploy Nutanix on bare-metal in AWS and then hibernate the cluster once configured, you would not pay ongoing bare-metal utilization costs. Instead, you’d pay for the time it takes to deploy and configure the Nutanix cluster, and then you’d only pay for the storage consumption once you hibernate. Therefore, significantly reducing cost. But, what good is a hybernated Nutanix cluster in AWS? Let’s take a look!

Watch this demonstration that captures the incredible integration of HYCU, Nutanix, and AWS.

As a bonus! Remember in my previous post, micro-seg-ment-tation, where I discussed the simplicity of isolating environments and securing east-west traffic based on application flow? You’ll see a great demo on Flow while I backup on-premise and recover in AWS. I’ve also recorded deploying workloads using Nutanix Calm, showing you the simplicity of deploying applications from a blueprint.

You won’t be disappoint[ed]!

Imagine delivering a solution that provides disaster recovery on demand, microsegmentation security polices, and application automation in under 30-minutes. Who else can do this? Is anyone up for the challenge?