Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Why the Type 2 hypervisor is making a comeback

Type 1 hypervisors used to provide a superior user experience, but advances in desktop and laptop technology have erased that advantage, and Type 2 client hypervisors are now back en vogue.

Type 2 client hypervisors went out of style a few years ago, but they might have just been ahead of the times.

A Type 2 hypervisor is installed on top of a host operating system, whereas a Type 1 hypervisor -- which is commonly referred to as a bare-metal hypervisor -- is installed directly on the hardware. Although Type 1 hypervisors typically use fewer resources and can support a larger number of virtual machines (VMs), Type 2 hypervisors provide the flexibility a lot of VDI shops are looking for nowadays, and more powerful desktops and laptops have improved the user experience of Type 2 hypervisors.

In 2004, VMware introduced a Type 2 hypervisor called VMware ACE. Like other client hypervisors at the time, it wasn’t very successful. There were niche use cases, but the hardware at the time simply couldn’t deliver a native experience when the VM was sharing resources with the host OS. ACE and other Type 2 hypervisors from vendors such as MokaFive, RingCube and Virtual Bridges gave way to Type 1 hypervisors, which are installed on bare metal.

VMware never got into the bare-metal hypervisor game. The company promised a product called CVP, but never delivered it. Several other companies that made Type 1 hypervisors came and went, and the two products that had any success -- Virtual Computer and Citrix XenClient -- consolidated when Citrix acquired Virtual Computer in 2012, and now Citrix is ending support for XenClient.

You can see the writing on the wall for the Type 1 client hypervisor.

Type 1 hypervisors ran into a similar problem as Type 2: There were only a limited number of use cases, so the demand never grew. The big drawback of Type 1 hypervisors is that updating your OS for all the new hardware means you’re always behind by six months. Plus, hardware is now capable enough that you can run a Type 2 hypervisor with no perceived lack of performance or user experience.

Challenge 1: Staying Current

The biggest challenge with Type 1 hypervisors is keeping them up to date. Because a Type 1 hypervisor is essentially a standalone Linux distribution, it requires all the maintenance you’d expect from any other OS, including support for all the different permutations of desktop and laptop hardware. New processors, chipsets and other hardware require OS modifications, followed by a lot of testing,  so the life of a Type 1 hypervisor developer is never dull.

The end result is that companies couldn’t always count on their bare-metal hypervisor to work on new hardware. Projects get delayed or killed entirely in favor of doing business as usual. After all, isn’t that what anything new is competing against -- business as usual?

Challenge 2: Hardware

Type 2 hypervisor products such as VMware ACE, which built upon VMware's Workstation cloud test-and-development environment, hold a distinct advantage over Type 1 hypervisors: They can run anywhere. Type 2 hypervisors rely on the host OS to provide the interface to the hardware, so all a Type 2 hypervisor developer has to do is make sure the hypervisor works on Windows, OS X or Linux. They still need to make updates, but can leave the hardware support up to the OS.

It might sound great, but in the first go-round, there wasn’t enough hardware on laptops and desktops to support VMs and still deliver a good user experience. Even if we decked out users’ laptops with enough memory and the highest-end Intel Core 2 Duo processor (yeah, it was that long ago), the experience they got was always a little weird because of all the emulated hardware. Type 1 hypervisors allowed you direct access to certain hardware components and could deliver an experience that Type 2 hypervisors couldn’t touch.

Type 2 is the better option

That’s all changed -- desktops and laptops have enough cheap memory to go around, plus processors are so powerful that if we have to carve off a bit to run a virtual machine, we can do so without affecting the user experience. In addition to that, Type 2 hypervisors have evolved to either provide access to the native hardware or to better emulate native hardware for the VMs, which means the user experience is better than ever. Add those to the classic benefits of a Type 2 hypervisor -- such as preserving the existing OS, installing on home or contractor computers and a normal user interface on the host -- and it’s easy to see how far we’ve come.

In the past, we had to choose between user experience (Type 1) and ease-of-use (Type 2), but today we no longer have to. You can see the writing on the wall for the Type 1 hypervisor. Citrix will shut down XenClient sales on October 1 this year, and VMware recently announced Horizon FLEX, which is basically the same exact thing as VMware ACE.

This is all happening at a time when the way that we traditionally manage Windows is being challenged. The idea of managing Windows via enterprise mobility management techniques is starting to take shape, and using client hypervisors to create dual-persona environments that are easier to manage is just one possibility in the future.

Next Steps

Background on how Type 2 hypervisors work

Choosing the right VMware Type 2 hypervisor

Differences between Type 1 and Type 2 hypervisors

Dig Deeper on Virtual desktop infrastructure and architecture