Migration to AHV Brownfield Considerations

As far as analysts and customers were concerned, hypervisors within the datacenter were a very mature segment that had been on cruise control for the past decade.

VMware was a leader for most of the 21st century. However, during the past decade, Nutanix has been developing AHV to meet customer demand and has made steady and noteworthy progress.

The demand for Nutanix AHV has surged since Broadcom's acquisition of VMware*, which has caused concern among its customers about the future of vSphere as a product. This set in motion expansion and growth for Nutanix AHV as customers began to consider hypervisor alternatives to Broadcom.

Adopting Nutanix AHV historically meant that you would also be purchasing new hardware and implementing our distributed AOS Storage fabric along with AHV as a complete private cloud solution.

This approach allowed for a highly polished solution that was designed to eliminate many customer pain points associated with legacy datacenter infrastructure deployments. Some organizations have significant investments in other datacenter products, such as industry-standard servers and other storage solutions.

What to do with existing infrastructure

What are your options for migration to Nutanix AHV if you can't abandon some or all of your existing infrastructure? This would be considered a brownfield migration as it repurposes existing infrastructure pieces and layers on Nutanix software.

There are currently two paths to re-utilizing existing infrastructure as a path to migrating to Nutanix AHV. The first involves support for VMware vSAN Ready Nodes. Nutanix can repurpose these existing vSAN Ready Nodes by integrating them with the Nutanix Cloud Infrastructure (NCI) and the AHV hypervisor.

The second is under the NCI Compute (NCI-C) offering that integrates with qualified storage partners. This allows AHV to utilize traditional external storage that organizations already have an investment in. The first partner to be certified with NCI-C is Dell PowerFlex.

In the following sections, I will discuss each of these brownfield options and provide a brief summary of what migrating workloads would look like after the conversion to AHV.

vSAN Ready Nodes

This approach aligns with modern infrastructure needs by leveraging existing hardware investments to support everyday workloads and emerging trends like AI and modern apps.

To better understand what this entails, let's examine the process of converting a cluster over to Nutanix and then migrating the workloads to the repurposed cluster now running AHV.

The first step is to identify the hardware to be utilized for the first AHV cluster. The cluster to be converted to AHV must be empty of workloads as it will be erased and Nutanix software will be installed on it.

If it has workloads, you will need to migrate them off before beginning the conversion. Once the new target cluster is empty, you can replace the vSphere and vSAN software with NCI and AHV.

This process is done with our foundation tool that enables you to select the target nodes and choose the AOS and AHV versions you want to deploy. The foundation automatically lays down the Nutanix software and creates the AHV cluster. When finished, you migrate workloads to the cluster.

The foundation process is relatively simple, and an average-sized cluster takes less than two hours to create. This process is required for each cluster that you convert from vSAN to AHV. Multiples can be done simultaneously or in a rolling fashion as infrastructure is made available in your migration plan.

NCI-C with external storage

This approach aligns well with organizations that have investments in traditional shared storage systems. While they may also utilize HCI architectures for some workloads, the external storage deployment is something they cannot move away from or choose not to at this time.

As mentioned above, the first of these is Dell PowerFlex. Initially, the PowerFlex 15G and 16G compute nodes are the only compute nodes on the certified list. These would be reimaged with AHV and AOS and connected to the existing PowerFlex storage deployment.

The Nutanix Foundation is used to image the nodes the same as with other deployments. In foundation, select the PowerFlex nodes that you want imaged and provide the necessary networking configuration details.

There is an option to configure them as compute nodes using external storage and selecting that will image them with the selected AHV and AOS versions on boot devices. Once the nodes are imaged, you are ready to login to Nutanix Prism Element for the new NCI-C cluster.

The nodes that were imaged have been used to create a new NCI-C cluster and upon first login, there are a couple of steps to finish the process. The first is to configure a storage network to connect to the PowerFlex storage and the second is to install the Storage Data Client (SDC) in AOS to communicate with the PowerFlex storage.

The SDC is downloaded from the Nutanix support portal and then provided to Nutanix Prism when you configure the connection. It will be installed on all of the nodes within the NCI-C cluster. You are now ready to start migrating workloads to the new AHV cluster utilizing PowerFlex as external storage.

Migrating workloads to the AHV clusters

Now that you have a target AHV cluster, you can configure your source and target destinations using the Nutanix Move tool. In this scenario, the source will be a VMware vCenter that is managing the cluster(s) that you are converting, and the target will be the AHV cluster.

Once those are configured, you can create migration groups. which are collections of VMs that will be replicated on the same schedule and can be either failed over together or in any granular configuration from a single VM to groups. The Nutanix Move tool allows up to 100 VMs to be grouped. For larger migrations, you would utilize multiple migration groups.

You can migrate the workloads when the target environment is ready, Move is configured, and replication is done. Start by selecting a migration plan or VMs. Move will shut down the source VMs and execute final replication of the changed data. The workloads are then powered up on the target AHV cluster and are ready.

Take a free test drive to learn, experience and explore hands-on NCI, Nutanix AHV and Nutanix Move.