Does the Hypervisor Still Matter?

Recently there has been a lot of talk about whether the hypervisor means anything to companies anymore. With the polar vortex blasting its icy air through Chicago, I figured I’d spend some of this enforced-indoor time to ponder on what I’m seeing in the enterprise space.

Whose Hypervisor Do You Have?

Do people still care which vendor’s product they are running for their virtual infrastructure? I’m going to go with yes. Sure, there may come a day in the next few years that it will not matter to a larger portion of the install base. From what I can currently see, the large majority of enterprise customers still rely on VMware as their primary hypervisor.

This means that when it comes to running their tier 1 applications and the majority of other applications, VMware is the choice. There is talk about Hyper-V being good enough now, but I see very little action in this space. There are a few smaller shops here and there that have switched or are testing Hyper-V, but that was usually because they think that Hyper-V was cheaper because they did not have to cut a PO for the purchase.

Management is not looking at the entire view of what it means to switch, and the difference is in managing and supporting such a change.

question

The Hypervisor: Time’s Running Out

I think in a few more years the hypervisor will be much less of a concern to more customers. As technology continues to advance and admin skill sets mature, the teams will be more less concerned about this layer in their stacks. Also, the tool sets for non-VMware hypervisors is still pretty immature; this will help adoption as third-party vendors make new tools or update existing ones to work with all hypervisors.

When it comes down to it many virtualization admins, the majority are focused on the nuts-and-bolts features of the hypervisor. They don’t care or don’t have time deal with the “fancy” features. I’m talking about the majority of people in the trenches. Sure there are still very large shops doing creative and cool things with KVM and other platforms, but these are still the exception.

Why It Won’t Matter

The expectation is that as non-VMware hypervisors continue to mature, their footprints in the marketplace will increase. I expect that as a customer I will be able to get the virtualization services that I need from any of the major three or four vendors. This will allow me to have choices based on price or OS.

I also expect that the focus will be on the layers above the hypervisors. So this would be the applications, services, or “cloud management” layer. As we automate and further abstract resources we will simply be concerned with the services rather than with which hypervisor is running on our servers. In some cases, if you adopt a product stack from a certain vendor, that may drive your choice, but I expect that these types of solutions will move further towards support multiple vendors.

Overall, I’m excited about what the next few years will bring in the virtualization space. Each vendor will advance their features and will enable new use cases. Don’t be scared about learning new things – that is how you will stay relevant!