Is application virtualization the answer for Windows Server 2003 users?

Is application virtualization the answer for Windows Server 2003 users?

Summary: AppZero is collaborating with Microsoft to persuade users of Windows Server 2003 to update before Microsoft stops supporting the operating system using AppZero's application virtualization technology.

TOPICS: Virtualization

Not too long ago AppZero chief executive Greg O'Connor explained to me that Microsoft and AppZero were working together to help companies still running windows applications on Windows Server 2003 to run their applications on a newer version of the operating system.

O'Connor pointed out that AppZero's "up level" app migration makes it possible to quickly and easily move applications from old, bare metal machines running outdated versions of the Windows OS to any server, anywhere — cloud or data center-running newer operating systems such as Windows Server 2008 R2 or Windows Server 2012.

What is the problem?

O'Connor offered the following statistics that were gathered from a number of sources to help me understand the industry pain that AppZero hopes to address:

  • 75 percent of the machines in the Fortune 5000 are Windows Server machines
  • 55 percent of the Window Servers in production are Windows Server 2003, totaling 12-13 million machines
  • July 14, 2015 is Microsoft's planned date to end support for Windows Server 2003
  • Less than 25 percent of the Fortune 5000 IT survey have a plan for WS2003 end of support

It's O'Connor's position that companies still using Windows 2003 Server should be making plans right now to migrate off of that operating system. He would point out that AppZero's application virtualization tools are a very good way to easily move on.

What is application virtualization?

As pointed out in my O'Reilly Media book, Virtualization: A Manager's Guide, server-side application virtualization offers a number of benefits including the following: 

  • Greater application isolation — Organizations need to make incompatible applications run side-by-side on the same system. This requirement can arise when the applications were purchased from third parties and require support of different versions of the same tools. Typically new versions of application development and run time tools replace older versions when they are installed. This can cause older applications to fail.

  • Operating system independence — Applications designed for an earlier version of an operation must be made to work with a newer version. Running applications in a virtual environment can significantly extend the life cycle of an application, giving the organization time to update the application or find a replacement for the application.

  • Improved availability — Applications provide critical functions and the organization will suffer great harm if those functions become unavailable for any reason. Applications can be made to fail over from one server to another.

  • Improved performance or scalability — Applications must perform faster or be able to service larger numbers of users. Server-based application virtualization products often include a workload management function, allowing the same application to automatically be started on multiple systems, to either improve application performance or allow more people to access the application simultaneously.

  • Cost reduction — It is necessary to reduce the cost for provisioning, installation, updating, and administrating applications. It is far easier and less costly to provision systems, install software, update software, and the like if it can be done from a central location. Once encapsulated, or placed in a virtual environment, applications can more easily be copied to remote systems or streamed to remote systems when required.

In this case, AppZero's tools address both application isolation and operating system independence.

Is application vitalization really a panacea?

As pointed out in my post "Is application virtualization the answer for users of XP?", the answer to the question "Is application virtualization a panacea for this problem?" is "no."

Application virtualization can make it possible for an older application to run on a newer operating system. The technology doesn't have the ability to address all compatibility issues. Here are a few situations that aren't helped by the addition of application virtualization:

  • The application provider's software license or terms and conditions prohibit the use of application virtualization. While the technology may work, customers would not be able to obtain support in this environment.

  • The application provider may not support the old application on the new operating system. If the application supplier still exists, it may require a version of software designed for and tested with the new operating system.

  • The application may require a specific device or set of devices that are not supported by the new operating system. A database application might be tied to a specific storage device or storage server that are not supported in the new environment.

  • The user interface will still look like it did under the older operating system and that may look strange in the new environment or not work well.

While application virtualization can be a wonderful solution and simplify the transition to a new operating system, it can't be considered a panacea. There are times that it would be wise to start over with tools and applications designed for the new operating system.

It may also be wise to consider moving from Windows Server 2003 to a different operating system entirely. This might mean migrating to a new set of applications or tools. It might also mean moving from a physical host to a cloud services environment. 

Topic: Virtualization


Daniel Kusnetzky, a reformed software engineer and product manager, founded Kusnetzky Group LLC in 2006. He's literally written the book on virtualization and often comments on cloud computing, mobility and systems software. In his spare time, he's also the managing partner of Lux Sonus LLC, an investment firm.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Application virtualization not all it's cracked up to be.

    Most application virtualization systems don't overcome all operating system dependencies, so it's not a perfect solution. It's fine for application isolation and portability, but compatibility is not something that app-virt should be considered for. If a business doesn't move forward with software upgrades, they aren't doing IT right.
    • This.

      A lack of software improvement isn't a sign of old technology, it's a notice of lazy developers.
      • Yup.

        It's also a sign of a software company that doesn't understand technology from a business standpoint - they can't just develop a single version of an application, then forget about R&D and switch all of their staff over to a support role forever.
      • ah now developers suck..

        ...of course. kill them all!!!
        breed a new kind! the one that will say every morning "hail bill" "hail steve"
      • Don't blame the devs

        Almost all of the software running on these old machines either have updated versions or they have gone EOF a long time ago. The problem is that software does not have rust or peeling paint, etc and so companies tend to ignore software and spend elsewhere.

        Blame management.
        Rann Xeroxx
    • Missing the point?

      Application virtualization isn't operating system virtualization. So it isn't meant to "overcome all operating system dependencies". If the software wouldn't run on the new operating system when installed normally, it probably wouldn't run in any kind of application virtualization container on that operating system. If there are exceptions, they must be very specialized. So I think the point of this article is that there are tons of applications out there, on Windows Server 2003 machines, which WOULD work on Windows Server 2012 machines if they could be reinstalled or moved. And then it presents this application virtualization technology as a way to move the applications when reinstallation is impossible or cost-prohibitive. Although that wasn't explained very well.

      It seems like a good idea if the main motivation is getting off of unsupported operating systems and aging hardware. I'm sure there are plenty of companies out there facing this sort of problem. Sitting around complaining about the causes and playing the blame game won't solve anything.
  • About panaceas . . .

    Of course virtualization is the answer sometimes. It's a great technology.

    "Is application vitalization really a panacea?"

    Probably not. Sure, it has some benefits, but describing it as a "panacea" is a whole different thing.

    There's very little in this world anymore that can be truly described as a "panacea," although ZDNet bloggers seem to latch onto that idea quite often.

    Tablets are somehow assumed solve all of the problems of notebooks and desktops, although that's never really been shown.

    The "cloud" has been assumed to solve all of the problems of local apps - although again, that has never really been proven.

    Bloggers seem to live in this fantasy world where everybody only uses 5 apps and the internet.

    Nothing's really as black and white as bloggers seem to want to paint it, and I'm sure the same is true for virtualization; it's going to depend on the needs of the business.
  • Seems like a solution to a non-existent problem

    When you start talking about LEGACY - you encounter the strong 'if it ain't broke - don't fix it" mentality. IT departments cringe at the thought of touching Legacy applications - so why would they be interested in touching a Legacy app to add this virtualization layer? They wouldn't and won't. The best strategy is to use the impending upgrade deadline as a lever to get programmers to update their app. Legacy costs - and adding a virtualization layer does little to change that dynamic.
    Roger Ramjet
    • Oftentimes not possible

      There are many reasons why it's not always possible to update an app. Sometimes the tools used are obsolete or there isn't simply the need to spend the money to do so.

      Additionally, I doubt there are many companies that are holding onto old hardware simply for that reason. It's costs way too much money to fix old machines. Where I work we have specified lifespans for hardware; if we can't move off an OS for whatever reason, we virtualize it.
  • no it's not...

    ...the solution is to totally abandon microsoft.
    • Business abandoning M$

      If only....
  • don't get the point of this article

    windows 2003 server isn't going to be running apps, unless it is a terminal server, one of only many server use cases.

    At best it fits that niche only. Article is of no use to an IIS .net server config.
    • no apps on 2k3?

      Hi there.

      Why would no applications be running on 2k3? And why is the article of no use to an IIS web server setup?

      Thank you.
      • I agree

        To me it seems obvious that there would be machines out there running important applications on 2K3. As long as the OS was supported and the machine was running smoothly, why would they change? I'd wager that there are still machines out there running Windows 2000. And that most of them are owned by banks.
        • I agree with CloudChronicle's questioning, I mean

          (See subject)