It’s rare for a business to get by anymore without running virtual machines, or VMs. Constant availability is the expectation, and complex tooling and confidential data have made virtual environments one of the only ways to achieve it. With the right hypervisor in place, anyone can accomplish work at any time, anywhere. Don’t have the software? Not a problem, access it remotely. Need to work with data that can’t leave the boundary of your organization? Easy, a virtual environment lets you slide into the organization without ever stepping foot in an office.
Despite the importance though, few companies are actually assessing their hypervisor options. More often than not, it’s a case of first click. Senior leaders may take the top recommendation from AI, search, peers or consultants — or just use what they’re used to — assuming relative parity across tools.
The wrong hypervisor won’t derail your business, but the right one can save you enough time and money to justify the extra attention.
Why trust Summit as an expert on hypervisor options?
Summit sells managed private and multi-tenant clouds, running on hypervisors. So why trust us, when we want to help you set up virtual environments? Well, the first is experience. Summit has been providing full-stack managed infrastructure for more than 20 years, and we do it really well. We run 22 data centers worldwide, and have a 100% uptime SLA. We create virtual environments for mass-market applications and heavily regulated industries, and we have a 92% satisfaction rate across it all.
Perhaps more importantly though, Summit is one of the only managed service providers not incentivized to give you a single answer. We are hypervisor agnostic — any kind of virtualization you want, you can get here.
Instead of recommending the platform that gives us the biggest cut, we work with you to understand what needs to be virtualized and what virtualization needs to accomplish. Then we recommend the best fit, and we help you build it. It seems simple, but most of our competitors are one-option shops. Whether you’re a nail or a screw or a staple, you’re getting the hammer — if the hammer in this metaphor is one particular hypervisor platform.
Hypervisor assessment criteria: What matters when choosing virtualization providers
There are a million-and-one technical differences, personal preferences and long-term biases that drive hypervisor debates and decision-making, but fundamentally, every option in this article works. When it comes down to impact on the business, however, there are four criteria that make the difference:
It’s easy to look at the licensing cost of each hypervisor and pick the lowest number, but that’s just one line in a much longer bill. Add the engineering hours to deploy and patch it, the backup and disaster recovery tooling it requires, and the hardware refresh cycle, and two platforms with similar sticker prices can land hundreds of thousands of dollars apart in total cost of ownership by the end of five years.
When we talk total cost of ownership, total cost of management is right in there. How much will it cost your team in terms of time and resources to keep the virtual machines running reliably? For many, the answer will be a straightforward “too much.” This is where a managed service provider like Summit can step in and handle updates, maintenance and up-time. A managed private cloud will cost you more on paper, but only because it externalizes all of the internal costs of self management. Don’t write off any options without knowing who will be running it over the long-term, and what those costs will look like.
If you are subject to HIPAA, SOC 2, PCI-DSS, or contractual data isolation requirements, that is an absolute standard you have to meet. Many options will be ruled out automatically, while others may need to thread a fine needle to stay compliant. Here, you’re looking not just for the option that saves you the most money, but the one that spares you the most fines.
Exit cost may have once been left out of the conversation when choosing hypervisors, but not anymore. When VMware abruptly changed plans and raised prices after being acquired by Broadcom, a lot of companies realized just how embedded they were in the platform, and just how much it would cost to leave for a more efficient option.
At Summit, we are always architecting virtual environments for portability. It’s a best practice, and it’s also in our best interest. Our competitive advantage is the ability to help you choose the right platform for your business. We want you to be able to move if that changes. Proprietary backup formats, custom networking configurations and platform-specific tooling can all add friction to future migrations. Before you commit, make sure you understand what it will take to leave.
Type 1 vs. Type 2 Virtualization
Type 1 and Type 2 describe where the hypervisor runs. They do not describe who owns the infrastructure, who manages it, or whether it is private or multi-tenant. VMware ESXi, Hyper-V, and KVM—the technology behind Proxmox—are Type 1 hypervisors commonly used for enterprise server workloads. Type 2 hypervisors run on a conventional operating system and are more commonly used for desktop development and testing.
For the rest of this article, Type 1 is already assumed. The more useful customer decision is the operating model supporting it.
Private vs. Multi-Tenant
Multi-tenant cloud uses shared physical infrastructure with logical separation between customers. It can support secure and regulated workloads when the required controls are in place.
Private cloud dedicates the underlying infrastructure to one organization and is often the better fit for predictable performance, licensing control, specialized hardware, customization, or contractual isolation. Neither model is automatically more secure; the right fit depends on the workload and its control requirements.
Comparing Type 1 Hypervisor Options: VMware, Hyper-V, & Proxmox
The Establishment: VMware
VMware’s vSphere platform has been, and is, one of the most mature and widely supported virtualization platforms. This is the most conventional choice, with the most robust third-party ecosystem. If you want a backup vendor, monitoring tool, or certified engineer, chances are you can find one that suits your fancy.
And yet the list of hypervisor options doesn’t stop here. Since being acquired by Broadcom, VMware has seen a surge of competition. The frenzy is driven, as it often is, by cost. The new ownership got rid of many products and effectively raised the prices of the rest. Now, everything is on a subscription basis, with minimum core counts. For enterprises running huge deployments, this is annoying, but not necessarily a dealbreaker. For smaller operations, paying more money for more licenses than you need is nonsensical.
None of this changes what VMware can do — it’s still an excellent hypervisor. But the math is shifting in favor of other options.
Total Cost of Ownership: VMware has the highest visible software cost of any of the solutions below, but that’s only part of the story. Licensing is expensive, full stop, but it can give you savings elsewhere
Management: Everyone in the industry agrees VMware is the easiest to manage. It’s well-designed for enterprises, and has the biggest pool of talent trained in using it.
Compliance: Compliance teams like VMware because they’re familiar with vetting it. It’s not necessarily more secure than the other options, and Hyper-V has easier documentation access, but it is the least risky in the eyes of most regulators.
Exit Cost: This is where the total cost of ownership spikes back up. There’s an incredibly high cost to exit. Even if you’re not relying heavily on customized components, people and processes will need to change.
Good for: Compliance-heavy workloads; organizations already using and staffed to support VMware.
The Up-and-Comer: Hyper-V
OK, it may not be fair to call a hypervisor built and backed by Microsoft an “up-and-comer,” but the truth is that Hyper-V was a bit of an afterthought for the past decade and change. It couldn’t match the features or ubiquity of VMware, and was thus ignored. Even Microsoft didn’t seem particularly dedicated to building it out and up.
All of that has changed in the past few years. Thanks to Microsoft’s investments — and the slew of people looking to leave the now-more-expensive VMware — Hyper-V has now mostly caught up to VMware’s capabilities. And for certain types of companies, there are other reasons to choose it.
Hyper-V is often attractive to organizations already invested in Windows Server and Microsoft administration. Licensing is included with Windows Server, which can provide a cost savings depending on how the physical host and Windows Server guest agreements are structured.
Hyper-V also fits teams that use PowerShell, Failover Clustering, System Center, Windows Admin Center, or Azure services. While not exclusive to Hyper-V, Azure Arc can also extend management and governance to supported on-premises and multicloud resources, acting as a single pane of glass for a company’s entire infrastructure. For those already in it, the interconnectedness of the Microsoft operating model can be a real advantage.
Hyper-V still has a smaller third-party ecosystem than VMware, but it is a practical alternative for organizations that already have the Microsoft skills, tooling, and licensing model in place.
Total Cost of Ownership: Hyper-V is included with Windows Server. For those with existing Microsoft agreements, there may be huge savings. For the rest of us, total cost will come down to host and guest licensing, management tooling, and support.
Management: Next to VMware, Hyper-V has the biggest ecosystem of third party support and managed service providers, making it easy to outsource or hire in-house support.
Compliance: While regulators may be less familiar, that’s changing. Hyper-V also has the most accessible documentation of any hypervisor. And, Microsoft has a deep history around compliance.
Exit Cost: Easy to move from private Hyper-V cloud to public Azure cloud. Moving to Hyper-V from VMware creates an opportunity to build for flexibility.
Good for: Businesses already running on Microsoft services, companies looking to reduce their VMware footprint while still using a “name brand” hypervisor.
The Open-Source Disruptor: Proxmox
For those looking to operate without VMware’s pricing restrictions and outside of the Microsoft Universe, Proxmox has quickly become the go-to option. It’s open-source, with no proprietary software layered on top (see below). Those who want technical support can get it via subscription. It’s also less expensive — no licensing costs, no surprise increases. For companies who aren’t already using Microsoft products, Proxmox is typically what we recommend.
The only real downside is the size of the ecosystem. VMware and Microsoft have larger pools of trained talent, certified integrations, and third-party tooling. The Proxmox community is growing, and growing fast, but it is still harder to hire someone with the skills to build and operate those environments. Working with a managed service provider like Summit can insulate you from some of those differences — it’s our job to recruit the best technical talent — but there may still be things that just don’t exist in the ecosystem yet.
For most businesses we talk to, Proxmox is attractive, but in a future sense, something to get to down the line. That leaves a lot of potential benefits untapped, while folks wait for a few marquee-name companies to take the leap first.
Total Cost of Ownership: The open source nature of Proxmox means no licensing costs are required, though there are still options to pay for support.
Management: Proxmox can’t match the number of developers and architects working in VMware or Hyper-V, but it can beat both on enthusiasm. For top talent though, Proxmox is a perk that allows them to work at the edges of the technology.
Compliance: While documentation may not be as readily available as it is for Hyper-V and VMware, Proxmox is more than capable of meeting regulations.
Exit Cost: The tool is open source, meaning no one is trying to trap you there. Though environments can end up so highly customized that only their architects can unwind them enough to move.
Good for: R&D workloads, SaaS-focused customers, organizations with a serious virtual footprint and the technological talent to realize benefits at scale.
Bonus Option: Kubernetes & OpenShift
Kubernetes is not a hypervisor, but it can be a way out.
Kubernetes allows users to deploy and run containerized applications consistently across supported infrastructure. Virtual or bare metal, the destination doesn’t matter as long as the applications are already containerized.
Kubernetes has no software license fee, but it does still have all of the other production costs: infrastructure, storage, networking, observability, security, backup, upgrades, and support. It also only works for applications that are or can be containerized. This is the main barrier. Some applications can’t be containerized, and others would require too much time, effort, and money to justify containerizing.
If you’re building new though? The benefits of Kubernetes make a compelling case for containerizing all new applications.
The Pros and Cons of OpenShift
For companies that do have containerized applications, OpenShift takes Kubernetes and packages it with additional platform capabilities and Red Hat support. That extra structure can simplify enterprise operations, but it also introduces subscription costs and a more opinionated operating model.
The right choice depends on whether the organization wants to assemble and support its own Kubernetes platform or purchase a more integrated one.
Total Cost of Ownership: Because Kubernetes and OpenShift are not hypervisors, it’s not quite fair to directly compare their total costs. That said, there are opportunities to save depending on the application, modernization effort, infrastructure, platform tooling, and support model.
Management: Kubernetes requires platform expertise or a qualified managed provider to run, and that may be harder to find than it is with the hypervisors. OpenShift provides additional tooling and vendor support but still requires ownership.
Compliance: Containers support regulated workloads, though compliance still depends on the full architecture, controls, data handling, and operating practices.
Exit Cost: Containerized applications are built for portability, and can offer more freedom to move across underlying infrastructures. The only exit costs come from data services, platform integrations, networking, and operational tooling.
Good for: New workloads. If you’re building something, containerizing it from the get-go is worth considering.
The Case for a Single Managed Infrastructure Partner
In a perfect world, each workload would dictate its own virtualization method. It wouldn’t be at all unusual for a company to run production line-of-business applications on a private Hyper-V or Proxmox cluster, a development and testing environment on multi-tenant cloud, a Kubernetes cluster for a customer-facing application, and a couple of Mac minis for the mobile team’s build pipeline. And we do work with companies like that! Because we’re hypervisor agnostic, we can manage multiple environments for them in our data centers while they run others elsewhere.
For companies that aren’t selling tech, however, the complexity of a system like that requires more attention than the marginal benefits could be worth. In those instances, we recommend an 80% rule. If a hypervisor is the right fit for 80% of your workloads, you’ll get more value out of using it for the other 20 as well.
If this all sounds like more than you want to manage — or more than you want to even think about — Summit gives you options. We don’t work for the hypervisors, we work for you. You choose the direction, we create the path. You can come to us purely for technical support, or for strategy, working together to find the optimal mix of hypervisors and design the VMs.
Ready to see what’s possible with today’s hypervisors?
Reach out to our team.