Virtualization Technology News and Information
Five Questions Desktop IT Managers Should Consider Before Adopting Desktop Virtualization

What do virtualization executives think about 2009?  A Series Exclusive.

Contributed by Dan McCall, President and CEO, Virtual Computer

Five Questions Desktop IT Managers Should Consider Before Adopting Desktop Virtualization

Riding a continuing wave of hardware cost savings, greater management efficiency, and improved security through server virtualization, IT managers have been quick to turn their attention to end-user desktops as the next frontier for virtualization.  There is little love among the IT community for traditional agent-based approaches to PC software delivery, patching, and security, creating a large window of opportunity for desktop virtualization to revolutionize PC management in the same way it already has server management.  Why then hasn’t desktop virtualization taken off at the same pace as server virtualization?  The answer is that while server virtualization and desktop virtualization share common technology underpinnings, the business drivers and implementation requirements for each are different.  Before desktop virtualization can be brought into the mainstream, the following fundamental questions must be answered in the minds of IT decision-makers:

1.  What are the real business drivers for desktop virtualization?

Few would argue that the overarching business driver for server virtualization is hardware cost reduction.  If you can realize hard cost savings on hardware, power, cooling, and other data center expenses by consolidating multiple physical servers into one, the business case comes into focus very quickly.  In contrast, the business drivers for desktop virtualization have less to do with hardware and infrastructure cost reduction and more with reducing PC management costs and improving security.  Cost savings from improving manageability, reliability, and security are less tangible than the hard cost savings of server consolidation.  However, done correctly, desktop virtualization will dramatically reduce the costs affiliated with user administration, hardware deployment, software deployment, application management, backup, archiving, recovery, service desk support, security management and IT administration.  A path must be provided for desktop virtualization adoption that has a small upfront investment, a straightforward implementation approach, and the option to migrate to virtual desktops at a comfortable pace without requiring a wholesale or “Greenfield type” upgrade. 

2.  Do virtual desktops need to run on a server?

Most of the early innovation in desktop virtualization has been centered on server-hosted desktop models.  However, this has been driven more by technology availability rather than an objective assessment of the market.  With laptops representing over 50 percent of personal computers sold today, a purely centralized computing model will not satisfy the needs of most organizations.  This is not to say that running desktops on a server is a bad idea.  Server-based computing, particularly server-based application delivery products like Citrix XenApp, have been used with great success.  Server-hosted desktops are a logical extension of this and we expect to see broad adoption of this technology in larger enterprises where the expense of server based computing can be more easily absorbed and larger IT teams are available to manage the infrastructure.  However, to achieve mainstream adoption by organizations of all sizes, desktop virtualization solutions must provide an option to execute virtual machines directly on existing PC hardware, including the loosely connected laptop devices that represent about half of corporate PC shipments today.

3.  What about PC-hosted virtual desktop solutions that already exist?

There are a number of PC-hosted virtualization products in the market today that are capable of running virtual machines locally on PC hardware.  VMware Player/ACE and Microsoft Enterprise Desktop Virtualization (formerly Kidaro) are notable examples.  These are good products with valid use cases such as desktop portability and Windows XP/Vista application capability.  However, because they have dependence on a non-virtualized host operating system, they don’t address the key business drivers highlighted above: manageability, reliability, and security.  Any reliability or security issues with the host operating system will be passed on to the virtual desktop.  We consider PC-hosted virtual desktop solutions the “1.0” products in this market and as such they have not achieved widespread adoption. It is only when you move to a bare-metal hypervisor architecture do you get the full benefits of virtualization and can solve the real problems facing PC configuration and management. 

4.  What will a mainstream desktop virtualization solution look like?

Given the growth of mobile computing when compared to stationary desktop computing, as well as the challenges of bringing server-based computing “down market” to smaller organizations, the model Virtual Computer is  pursuing with NxTop, is one that offers centralized management of virtual machines combined with simple, reliable, and secure execution on a virtualization engine running directly on the client device. We call this centralized management with distributed execution, which is in contrast to products such as VMware’s VDI that promotes centralized management and centralized execution.  Using virtualization in a distributed fashion allows companies of all sizes to replace traditionally installed desktop operating systems with virtual desktops using the same hardware profile and cost model they are operating under today.  Mainstream desktop virtualization is not a new model of managing computers, it is just a new computer that needs to be managed.  When architected correctly, the management of this new virtual computer can be simplified and greatly improved by eliminating the agent based approach in use today.

5.  Will the end-users be happy with the solution?

End-users are by nature resistant to change. For desktop virtualization to be accepted by end-users, it needs to deliver “perks” to users – not just to IT.  We see a number of benefits to end-users that will have them asking their IT departments to adopt this technology. Some of these benefits include:

  • The ability to run a personal operating system that can be customized and side by side with a standard corporate desktop, that IT can keep under tight control.
  • Peace of mind that sensitive data is safe and secure and that their entire computing environment can be completely recovered in the event their laptop is lost or stolen or damaged.
  • Faster PC boot and shut down times.
  • Non intrusive patching that cannot cause their computer to fail once the patch is applied.

This is only the beginning. As this new model of personal computing emerges, IT managers and their end-users will create new usage scenarios to take advantage of the virtual hardware in the same way that they have helped evolve server virtualization platforms into the mission critical infrastructure they have become today.

About the Author

Dan McCall, President and CEO

Dan is a high-technology veteran with over 25 years of experience. Immediately prior to Virtual Computer, Dan was a co-founder of Guardent, a managed security services provider, which was sold to VeriSign in 2004.

At Guardent, Dan led product management and marketing to define overall company and service-line strategy and to grow company revenues. Dan was also responsible for 24x7 operations at Guardent, which became the backbone of VeriSign’s MSS business. After Guardent’s acquisition, Dan became VeriSign’s Vice President of Corporate Development and worked closely with senior management and multiple business units in the area of strategic planning.

Before Guardent and VeriSign, Dan was Vice President of Worldwide Marketing at i-Cube and led worldwide marketing and project management for Shiva Corporation, where he helped introduce Shiva’s first VPN technology.

Dan started his career in 1983 as a software engineer and, in addition to the above experience, he has held engineering management positions at Encore Computer and Wang Laboratories.

Dan is proud to have been the recipient of the prestigious Ernst and Young 2002 Entrepreneur of the Year award and has been quoted in numerous industry and business publications, including Business Week, Wall Street Journal, New York Times, Boston Globe, Information Week and eWeek.

Dan received a Bachelor of Science degree in engineering and computer science from the University of Connecticut, graduating Magna *** Laude.

Published Tuesday, December 16, 2008 7:06 AM by David Marshall
Filed under:
There are no comments for this post.
To post a comment, you must be a registered user. Registration is free and easy! Sign up now!
<December 2008>