KVM, the kernel based virtual machine software that is a standard part of the Linux kernel and reached a point in which it is beginning to stand out from the crowd. I'm hearing from more and more suppliers of hardware, software and services that are mentioning their support of this technology. The topic has come up in so many conversations recently that it appears to have become a trend. My recent discussion with Arsalan Farooq, CEO & Founder of Convirture, was really interesting (Convirture is doing many innovative things) and served to crystallize my thoughts on KVM.
Here are some reasons that KVM is now taking its place on virtualization's world stage:
- KVM's performance, reliability and manageability are at levels that make it good enough for most tasks. It can compete effectively with VMware's ESX, Microsoft's Hyper-V, and Citrix's XenServer.
- The fact that it is open source software and can be used as needed by organizations rather than used only as suppliers allow makes it attractive to service providers and others.
- The pricing model does not amount to a "hypervisor tax" on an organization's processing.
- The emergence of the open virtualization alliance demonstrates that there is a healthy ecosystem supporting this technology. (see posts by my colleagues Paula Rooney (Red Hat-IBM pact, OVA launch will drive more KVM use in enterprise and Red Hat declares war against VMware on cloud front) and Ken Hess (OVA: Virtualization's Good Eggs) for their thoughts on the Open Virtualization Alliance)
If I review many of the major tipping points in the world of IT including the move to adopt PCs, client/server computing, relational databases, UNIX and Linux, I see a common thread. When the technology gets to the point that it is "good enough" and the pricing and conditions are more attractive than well established alternatives, the technology takes off.
Is now the time for KVM?