What do Virtualization and Cloud executives think about 2011?  Find out in this VMblog.com series exclusive.

Wiping the dust off the crystal ball we keep locked away in a special, darkened room here at Quest and peering into the not-so-distant future that is 2011, what do we see for desktop virtualization? Well, a few things:

1) Cost reduction. To date, VDI adoption has been most aggressive in areas where it is driven by requirements other than cost reduction. For example, hosted Windows XP and 7 desktops have been adopted where security or data sovereignty or device independence are such imperatives for the business that they are willing to pay higher CapEx costs than traditional desktop deployments. This is evidenced by the fact that the strongest vertical markets where we have seen early adoption are defense, healthcare and finance. Interestingly, education has been another early adopter, partly because education licensing and hardware discounts skew the economics in favor of desktop virtualization more than in the commercial sector, and partly because education establishments have students who bring their own devices outside the management of traditional IT. Users in academic organizations require more flexibility than a traditional desktop can deliver.

For desktop virtualization to become more widely adopted requires a significant reduction in the CapEx cost far below that of a physical desktop so that it becomes a serious consideration for all organizations, not just those with specific needs. To move the CapEx needle significantly multiple approaches are needed: (a) greater use of Terminal Server (more of that in a moment), (b) optimization around storage, (c) higher density of user VMs inside each virtualization host and, (d) a smarter approach for offline and mobile users than check-in/check-out of an entire VM from the datacenter. We expect to see major progress in all these areas in 2011.

2) Terminal Server/Session Host Resurgence. Throughout 2011 we encountered failed VDI pilots that stalled on economic grounds because the CapEx cost per user was simply too high compared to a traditional PC. Despite the fact it has been available for 12 years, there are far too many IT people who have simply never heard of Terminal Server, know renamed as Session Host. The fact is that 80-90% of the technology in VDI is the same as that used in Terminal Server and there's a simple, low cost way to deploy applications and desktops to task workers at a fraction of the cost of VDI using almost identical technologies.

There are at least two reasons for this ignorance: (a) VMware has driven a lot of awareness of VDI as complete virtualization of the desktop OS, ignoring other approaches, (b) the majority of Citrix's revenue comes from the enhancement of Terminal Server, but few people are aware that XenApp (formerly MetaFrame and Presentation Server) are extensions for this underlying nameless product platform.

As the market becomes more educated about desktop virtualization in general, not just VDI, we expect to see a lot more organizations realize that the needs of their task workers are met at a lower cost with Terminal Server than VDI, and that they need to blend these technologies to achieve the greatest cost savings. Of course, there are situations where Terminal Server won't work - application compatibility can be an issue, and support for some devices attached to user endpoints requires VDI - but it's perfect for task workers and office-based workers and Gartner has documented the cost savings over traditional physical desktops at length.

3) Hyper-V expanding market share in desktop virtualization. While ESX/ESXi is established as the king of server virtualization today, the features required for desktop virtualization are different and we are seeing much wider openness to evaluation of Hyper-V. Numerous customers we have worked with in 2010 have built financial models showing Hyper-V as the most cost-effective choice. With the imminent release of Service Pack 1 for Windows 7 and Windows Server 2008 R2 and Hyper-V Server R2, the Hyper-V platform is about to get a density increase that will deliver further cost savings. Coupled with the availability of RemoteFX on Hyper-V only, we are expecting a further shift in momentum towards Hyper-V in 2011.

4) Growing diversity and ubiquity in access devices. The iPad showed the world how far mobile device technology has come in 2010 and we're looking forward to another wave of new tablet devices in 2011. Anyone who traditionally would have carried a clipboard and used thin clients or PC hot desks to do their jobs is looking at tablets - the biggest demand is coming from hospitals, manufacturing and healthcare. We're also seeing a lot of demand in academia where students are trying them as laptop alternatives in campuses where desktops are available virtually. Another big development will be the inclusion of decoder chips for protocols like RemoteFX in TVs and monitors, essentially embedding a thin client into the display hardware, and with a new breed of super-low-cost thin clients capable of rapid graphics throughput coming to market we're going to see virtual Windows desktops and applications available in more places and on more devices than every before.

5) Cloud. "Cloud" has to be the most nebulous term of 2010, appearing everywhere and used in just about every possible context! In the world of desktop virtualization it primarily refers to virtual desktops being served from a co-located service provider's datacenter ("the cloud") rather than an on-premise one. This is very much Infrastructure-as-a-Service rather than Platform-as-a-Service, but it is early days either way. The biggest challenges are: (1) extending an enterprise authentication domain into the cloud, (2) effective diagnostics and self-service, and (3) licensing.

The diagnostics piece is particularly important since the cloud business model is built on ultimate administrative effectiveness and scalability, so a drill-down dashboard showing the current state of all virtual desktops with rapid identification of the root cause is essential.

Licensing has not yet caught up with the idea that the entire user desktop and applications can be delivered on-demand, as a service, paid for on a monthly or quarterly basis with no lock-in or perpetual contract. The momentum behind Cloud thinking is immense, and with the huge capacity and growing network bandwidth available at ever lower prices, customer demand will surely drive the pace and acceptance of desktops in the cloud and changes to licensing during 2011.

My time with the crystal ball is up and somebody else is knocking on the door of this hallowed space so I had better wrap up. 2010 might not have been the break-out year for desktop virtualization that many predicted, but nevertheless we saw remarkable changes and technology introductions. 2011 is shaping-up to see the industry take advantage of them, add more, and spur the pace of adoption.

About the Author

Jon Rolls is vice president of product management for the Quest Desktop Virtualization Group. Jon joined the Quest family in 2004, and brings 15 years of software industry and Windows management experience, including product management roles with Citrix, ScriptLogic, and virtualization and web acceleration startups.

Jon has a Bachelor of Science degree in pure mathematics from the University of Warwick, England.