Citrix Systems, EMC's VMware unit, Microsoft, Symantec, and other big dogs have shelled out more than US$2 billion on acquisitions related todesktop virtualization in less than two years. But now that they've got the technology, it's not at all clear that customers will buy it.
Indeed, analysts and surveys of big IT buyers point to different conclusions about corporate appetite for the technology.
"Desktop computing, as it is practiced in enterprises today, is broken. Windows desktops are not secure. They are plagued by email-borne viruses and other malware. For this and other reasons, they are too costly to manage," says Rachel Chalmers of the research firm 451 Group.
Chalmers cites a survey of 376 IT buyers by ChangeWave Research showing that 60 per cent already have budgets earmarked for virtual desktop technologies.
Desktop virtualization moves the actual desktop to the datacenter and provisions a copy to the user as needed. The promise for IT is that users can no longer mess up the desktop image that IT has decided is stable and authorized. Desktop virtualization also promises to let IT separate the various OS and application layers, thus making it easier to protect the core OS from other components and to rebuild any compromise pieces on the fly.
Is desktop virtualization's promise overstated?
Wait a minute. I certainly won't argue that the Windows desktop isn't broken, but is it really accurate to imply that desktop (remember, we're not talking about servers here) virtualization can fix all those ills? I'm not so sure. And neither are buyers.
Just last month, Intel released a survey that paints a somewhat less bullish picture than ChangeWave's. According to that survey, 39 per cent of the enterprises have a current deployment of desktop virtualization, but the number of companies doing "broad deployments" of server virtualization and related technologies such as application streaming, was in single digits.
And then there's a recent survey by our sister publication CIO, which found that 25 per cent of enterprises were using desktop virtualization and another 13 per cent planned to do so within a year. But 21 per cent said it would be one to three years before they deployed, and 37 per cent said they were not interested.
Unclear purpose, emerging alternatives weaken the case
"Desktop virtualization's greatest obstacle is the clarity of the business case," Burton Group analyst Chris Wolf told CIO. "Until the technology matures, you're not going to see the 12- to 18-month ROI that's common with server virtualization today. That's why I've seen enterprises willing to dip their toe in the water but not quite ready to jump in feet first."
Adding to the uncertainty are the multiple flavors of desktop virtualization, which Chalmers categorizes this way: client-hosted, device-hosted, server-hosted, and cloud-hosted. "We also recognize four closely related technologies: OS streaming, application streaming, OS-hosted virtualization, and terminal services," she says.
Chalmers adds that other mainstream computing companies, including CA, Hewlett-Packard, and IBM, whose core business includes managing traditionally configured and provisioned Windows desktops, are responding to the "threat" of desktop virtualization by building, as opposed to buying, their own solutions.
It's worth remembering that virtualization of all types isn't really a new idea. "Despite the appearance of rising to stardom almost overnight, virtualization is in fact a technology that dates back to the earliest days of computing and is really an ongoing phenomenon with an evolving role in corporate IT infrastructure," says Forrester analyst Galen Schreck.
Old or new, there's plenty of smoke here. But how much fire, and by that I mean profit for vendors and value for buyers, is a lot less clear.