Against the backdrop of an aggressive upgrade marketing push for Windows 7, it is still a somewhat remarkable statistic to note that essentially 50 percent of PCs still use Windows XP as their operating system. While adoption and migration is certainly on the rise, the large XP contingent begs the question: Why have they not converted from a 10-year-old operating system, given all the capabilities of Windows 7? And, equally important — with Windows 8 looming on the horizon — what should businesses do in terms of taking the upgrade plunge? When? And how?

The Wait for 8

At the earliest, Windows 8 is due to arrive in late 2012. Early previews indicate that there will be evolutionary improvements to Windows 7, but nowhere near the magnitude of changes implemented with WindowsXP.

The Windows 7 features provide a significantly better computing experience for both the user and the IT manager: support for more powerful 64-bit processors, a sleek interface with personalisation capability, better multimedia support, improved security, integration with management tools and improved support for peripheral devices. Since its introduction in October 2009, Windows 7 has proven to be a stable platform for business users and a vast improvement over its predecessor, Windows Vista.

The biggest barriers to Windows 7 (or 8) migration are complexity, cost and the fear that business operations will be disrupted. To plan and execute an upgrade as significant as Windows 7 is no small feat. It requires a concerted planning effort and a clear understanding of the associated expenses. Then there are the implementation challenges such as application compatibility issues, managing personal profiles, and new requirements, such as the increasing use of social media in a business framework.

It’s also a huge time investment — particularly as the scale and sprawl of the computer system/network increases. And it’s not just an enterprise problem: Smaller businesses have all of the same technical problems and headaches that an enterprise IT department does, only with fewer resources.

When evaluating options for an OS upgrade, there are really three key questions that define the challenge:

  • Complexity: How hard is it to perform from a technical point of view?
  • Cost: How much is the hard dollar cost for hardware and software – and what are the recurring soft-dollar costs in terms of IT support?
  • Maintenance: How easy is the new solution to maintain?

Among all the different products and vendors who provide upgrade solutions, there are basically only three core approaches to managing an OS upgrade:

  • Physically migrate the new OS on to each PC. This approach certainly gives a lot of direct control over the upgrade process to the IT administrator, but often at a prohibitive cost. Each machine has to be individually backed up, re-imaged, tested and then put back into the user’s hands. This becomes an increasingly expensive and time-consuming process when multiple offices (or even global operations) are concerned. Against all three of the key criteria (complexity, cost, maintenance), this approach only really presents itself as feasible for businesses with massive IT support capabilities and the strategic need to physically manage every PC.
  • Deploy a one-to-one Virtual Desktop Infrastructure. At first glance, because thin clients are cheaper than full-function PCs, the lower hardware cost should drive down the costs in this model…and the ability of a remote system administrator to re-initialise a failed virtual client should be cheaper than a hands-on desktop management process. In the real world, however, the results of this approach are often pretty similar to that of a one-on-one upgrade because even though you have removed the endpoint PC, you must still manage individual OS images for every user. The complexity of managing one-to-one VDI environments can often be underestimated, taking up more IT support time — not less.
  • Deploy a terminal services model optimised for multi-user environments. In this approach, user devices access individual desktop environments running within isolated computing sessions in a shared server host (up to 100 seats per host, in some instances). Simply put, this approach circumnavigates many of the challenges of traditional OS upgrade by virtualizing the upgrade. In short, it streams Windows 7 to the desktop to create a nearly identical user experience without the need to actually upgrade. It is a ‘virtual’ Windows 7 environment.

Microsoft has incorporated several new features into Windows Server 2008 R2 to provide a seamless Windows 7 experience to end users: broad business application compatibility, better integration of server-based applications with the user’s desktop, improved personalization, and powerful management tools.

As a result, administrators can begin to realise the promise and benefits of client virtualisation without having to deal with the managerial complications of a 1:1 VDI model or cost of a PC by PC approach.

This approach, if implemented properly, can result in significantly lower per-seat acquisition costs, as well as lower long-term maintenance and support costs — which means the ability to do a one-time application deployment for all users, improved support, centralised security and centralised data backup. Installing virtual desktop soft clients onto the old XP systems can also mitigate the high cost of migration. For those who have been contemplating both a move to Windows 7 and a move to virtual desktops, it is highly beneficial to combine these rollouts into one effort.

Of course, before undertaking any OS upgrade, it is important to undertake full compatibility and network stress testing to assure the terminal services-based virtual desktop environment can deliver the performance and benefits designed by the administrator.

The benefits of Windows 7 in terms of performance, rich functionality and ease of use are all powerful business reasons to consider an upgrade. Coupled with the means to achieving migration without the migraine via a virtual upgrade strategy – the path is very much cleared. There’s no longer a reason to ‘wait till 8’ and every reason to migrate to 7.