Many people (myself included) have said that 2010 will be the "year of the bare-metal client hypervisor." In other words, next year is when we'll finally see products from mainstream vendors and customers really starting to adopt them.
A client hypervisor, if you're not familiar with the concept, is essentially a server hypervisor that lets one or more virtual machines run on a desktop or laptop computer.
"But wait," you might ask, "don't products like Microsoft Virtual PC and VMware Workstation already let users run multiple VMs on desktops and laptops?"
Yes, they do, but these products are known as "Type 2" virtual environments, where they let VMs run on top of existing applications.
In 2009, two startup companies released bare-metal client hypervisors -- Neocleus and Virtual Computer. This is cool because it means that this stuff is actually shipping today. Unfortunately, not too many people have heard of either of these companies, and so most people are waiting to see what the big companies will do.
As I'm writing this in November 2009, Citrix is getting close to releasing the beta version of a bare-metal client hypervisor known as XenClient. (This was previously known as "Project Independence" if you'd like to Google it for more details.) And VMware has promised to release in the first half of next year its own bare-metal client hypervisor, which is currently known as VMware CVP, for "client virtualization platform."
So what's the big deal? I've been playing with these things in one form or another for a while now, and there are some key differences between client hypervisors and server hypervisors.
First of all, the motivations behind these two technologies are completely different. Server hypervisors are designed to make VMs portable and increasing the utilization of physical hardware. Client hypervisors are intended to increase the manageability of the client device and improve security by separating work and personal VMs.
Server hypervisors have to be tuned for maximum simultaneous network, processor and disk I/O utilization. Client hypervisors have to be tuned for graphics, multimedia and wireless connectivity.
Server hypervisors only have to run on a narrow set of different preapproved hardware models. Client hypervisors should (ideally) run on just about anything.
Server hypervisors are the dream of Intel because they push the processing capabilities to the limit, driving everyone to buy the latest and greatest. Client hypervisors -- well, I guess that's one thing that's the same for both!
The bottom line is that even though they're both called "Type 1" or "bare-metal hypervisors," there are some philosophical differences in how each came to be. (This could help explain why it has taken over five years to extend the Type 1 hypervisor concept from the server to the desktop.)
But I promise that 2010 will be the year that you start to see bare-metal client hypervisors in the mainstream, so you should think now about how these might help you your own environment.
|ABOUT THE AUTHOR:|
|Brian Madden, Independent
Industry Analyst and Blogger
Brian Madden is known throughout the world as an opinionated, supertechnical, fiercely independent desktop virtualization expert. He has written several books and over 1,000 articles about desktop and application virtualization. Madden's blog, BrianMadden.com, receives millions of visitors per year and is a leading source for conversation, debate and discourse about the application and desktop virtualization industry. He is also the creator of BriForum, the premier independent application delivery technical conference.