As I wrote in my inaugural predictions post, I am convinced that this year will see significant growth in the virtualization space. The magic here is to understand who would come out the winner. The trick about virtualization is that whoever runs the virtual machine monitor or VMM, is able to dictate how the guest operating systems will be able to behave. Redmond has a VMM capable OS (I think), and from all the rumblings I hear about the Redmond-Novell deal, there seems to be a tussle about who will be occupying the VMM space. Both of them think they will be occupying that space with the other kowtowing to the VMM. Therein lies the nub – the one that controls the VMM controls the system and wins. There will be a duel between these two and the corporate blood to be spilt, I suspect, would not be in Novell’s favour.
With that then, I feel RHEL5, when it goes GA, will be in a position to occupy that most critical of spaces. And the whole world would be the better for it. And, no, I am not saying this because I work for Red Hat, but because if you look at the dynamics of the marketplace, albeit the virtualization marketplace, the spot that should not fall into the arms of a convicted monopolist is the VMM space. Novell is clearly not in the running on this, given their death pact, and I do not see anyone else stepping up to the plate other than Red Hat and RHEL5.
Why would the VMM space be so important? That is the spot which touches the raw hardware at the bottom level, and at the top level, exposes the requisite APIs and objects for the guest operating system to work with. If the monopolist were occupying that space, they can then choose what is visible at the top level and to whom – not unilke the DR-DOS fiasco which suggested via the “env” command that running DR DOS instead of MS-DOS or PC-DOS was not OK to run Windows 286/386.