Citrix Looks Beyond VDI
Last week's Virtually Speaking looked at the virtualization vendors' growing interest in the virtual desktop. I questioned whether the virtual desktop will have its due in 2010. Not surprisingly, Citrix believes the the answer to be a firm "yes."Virtually Speaking: Within two years, 60 percent of all desktops will be virtualized, according to Gartner. Citrix has six avenues to make this possible.
Last Thursday, I met with Citrix CTO Harry Labana to discuss the company's plans for increasing its virtual desktop footprint. For some time now, Citrix has believed the virtualization will drive client devices in the not-so-distant future, and it is moving actively on that vision.
Citrix is not alone in this belief. Research firm Gartner predicts that "60 percent of corporate desktops will be virtualized all or in part by 2012."
Citrix currently claims 20,000 end users taking advantage of the six virtual desktop options that make up its FlexCast technology. Labana emphasized that the oft-cited virtual desktop encompasses far more than virtual desktop infrastructure (VDI) a desktop operating system that is in actuality a virtual machine (VM) hosted from a centralized server.
Unfortunately, VDI and desktop virtualization are often used interchangeably, and thus VDI is overused. At times, the virtual desktop isn't actually a desktop at all, but a client device. Note that the 60 percent that Gartner cited applies not just to VDI but to all types of desktop virtualization. Also interesting is that Citrix is talking about more than just desktops. Its technology is meant for smartphones and netbooks as well as desktops and laptops.
For highly mobile users, where supporting everything on the client is preferable, a client-side compute model is the way to go. The three options are:
- Local, VM-based desktops (offline mode), like XenClient
- Virtual apps that are delivered to installed desktops
- Local, streamed desktops (i.e., diskless machines booting from a single image)
For task-oriented workers, three other options may be more appropriate:
- Hosted blade PC desktops, machines cannot have a hypervisor
- Hosted virtual-based desktops the technology more commonly known as VDI
- Hosted shared desktops !51 the technology more commonly known as terminal services
Note that all of these options apply to end user. Ideally, a company will evaluate its workers needs and deploy multiple solutions accordingly.
Ideally, the technological choices will also be based on the drivers, whether they be business, technology or end user, Labana said. Users, for example, generally want more control or choice (e.g., the CEO wants a Mac), whereas security would be a business or IT driver.
One of the most frequent arguments against desktop virtualization is that it does not save money up front. Labana did not refute this assertion, particularly in relation to VDI, but he noted that desktop virtualization makes for a more agile enterprise, which facilitates lower costs elsewhere as well as growth.
In today's economic climate, cost savings is often a primary decision driver. In its absence, as is the case for the virtual desktop, an catalyst is needed. Labana said he believes Windows 7 may be just that.
Many enterprises passed on upgrading to Vista, and thus they are now ready for an operating system upgrade. In many cases they are also in need of a hardware upgrade, as what works fine for XP may not be suitable for Windows 7.
Desktop virtualization will make this sunk cost both easier to swallow and sell. The central control that comes with desktop virtualization means IT department have more control. This translates into increased security, more streamlined patching and updating, and easier management. All of which results in a lower total cost of ownership something that even the most technologically averse CFO can get on board with.
Amy Newman is the managing editor of ServerWatch. She has been covering virtualization since 2001, and is the coauthor of Practical virtualization Solutions.