We’re back with another “Feature in Five,” our byte-sized videos of industry-leading VMware product solutions, all in—you guessed it—five short minutes! Last time, we video-profiled vSAN Integrated vSphere Update Manager (VUM); in this edition, we’ll take a closer look at stretched software-defined data centers (SDDC) on VMware Cloud on AWS (VMC).
VMware Cloud on AWS Stretched SDDC
To begin, VMware Cloud on AWS is simple with vSAN. Even if you know very little about vSAN stretched clustering, the in-platform wizard walks through the process, answers all questions and makes it impossible to error while still giving you an element of control. From the VMware Cloud on AWS portal, you can see what SDDCs are already configured. Most SDDCs are standard multi-host SDDCs, which means all of the hosts are in a single Amazon Availability Zone (AZ). But we’re here to talk through the configuration of a stretched SDDC. So, let’s get to it.
With a stretched SDDC, multiple AZs will be consumed. To start, choose the deployment option; in this case we copy a Multi-Host Stretched Cluster. From there, give your cluster a name, then relay how many hosts you would like to configure. Next, you will be asked what AWS account and which AZ you want to deploy in. You then must select which Amazon virtual private cloud you would like to connect.
When you select the VPC, you are actually selecting what you want to connect to the logical workload network. This can be EC2 instances or Amazon services in conjunction with existing workloads, COTS (commercial off-the-shelf software), GOTS (government off-the-shelf software), or even virtual machines (VMs) inside vSphere. Select whatever VPC those services reside in to avoid the transitive routing through Amazon. This ensures they are all in the same subnet and lets you control costs to avoid any egress fees.
Now, the VMware Cloud on AWS wizard will tell you at least that they need to have a least one subnet. And since this is a stretched cluster, you will need to select a second one as well. On the backend, by selecting those two AZs you have designated the preferred and nonpreferred fault domains for the stretched clusters. The remaining availability, if applicable, will deploy the witness host into whichever one is not selected.
Finally, you will be asked for a management network. This means an IP address to be used for vCenter, NSX controllers, vSphere hosts, all the other platforms if you want to link back on-premises. From there, acknowledge the changes, and hit go! This process will take two hours to deploy and build out the environment.
And that’s how you create a stretched SDDC on VMware Cloud on AWS! This process plans for resiliency and failures, and vSAN takes care of the availability. If workloads crash or a cluster crashes, vSphere High Availability detects the crashed workloads and restarts them in the surviving AZ. Once the cluster is up and running the compute policies kick back into place, automatically moving the workload into place and re-establishing High Availability. It’s an easy way to increase your Availability on VMC. Learn more about Stretched SDDC in just five minutes by watching our video!