In spite of currently having a very small user base, Kernel-based Virtual Machine (KVM) is getting quite a bit of attention lately from the press and the virtualization market.
VMware still soundly dominates the market as well as the majority of the media discussions, but Microsoft Hyper-V is starting to pick up steam in both areas. Xen, on the other hand, has been around since 2003, and companies like Citrix and Oracle hope to make great strides leveraging this technology for their own purposes. But KVM is the newcomer here, and it has a long stretch of road ahead of it to catch up with these other technologies. It does have one thing going for it, however: The KVM technology is integrated into the Linux kernel.
Red Hat continues to be the main champion and major contributor of the commercial implementation of KVM, but it could be getting some help from some very unfriendly places.
Believe it or not, it looks as though Novell is now researching the creation of yet another virtualization platform. No, this one isn't based on Xen. Instead, the software company appears to be jumping on the KVM bandwagon and is looking to build a new open source project based on KVM.
Red Hat was the first to stick its neck out there in June 2008 when it announced that as part of its future virtualization strategic initiative, the company was going to throw its support and credibility behind the little-known KVM technology. It followed that announcement three months later with another statement, this time backing up its position with a $107 million acquisition of Qumranet, the company that originated the KVM technology. After that, Red Hat started down a path of shifting its enterprise software ecosystem from Xen to KVM. And in late 2009, Red Hat pushed that transition forward with the release of Red Hat Enterprise Linux 5.4, making KVM its preferred virtualization platform of choice over Xen.
Virtualized environments often impose significant performance penalties against a given workload when compared to native "bare-metal" equivalents. This project is motivated by the belief that it doesn't necessarily have to be this way, nor do we need exotic hardware to achieve it. AlacrityVM demonstrates that most of the existing performance bottlenecks in today's system are simply the result of suboptimal software stacks. By systematically identifying and fixing the weak links in the guest/hypervisor equation, near native performance from a virtualized environment is realistically achievable.
The project is still in the early stages of development, and it plans to support all of the basic features found in the upstream KVM project. The group also aims to add other new features, such as the ability to express real-time constraints, network QOS, virtual file systems, accelerated disk I/O, and accelerated Microsoft Windows guest support.
So if Novell continues to follow its open source rival Red Hat down the KVM virtualization path, what does that mean for Novell customers currently running Xen? And what, if anything, does that mean for future Xen support coming out of Novell?