Just because you can, doesn't mean you should virtualize the entire desktop

This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter's approach.

Although virtual desktop infrastructure (VDI) is well recognized, up until now, the technology has over-promised and under-delivered. Many IT departments, lured by better resource utilization and cost savings, have had their fingers burnt as a consequence of adopting a 'lock stock and barrel' approach to VDI projects. For many companies, the project has actually negatively affected users' ability to do their jobs, in turn resulting in outright rejection of the implementation.

[RELATED: The growing importance and network impact of VDI]

The error has lied in taking an IT-centric approach to VDI. Evidence suggests that a user-led approach to VDI implementation is more successful. This requires a deeper analysis of how users do their jobs and the impact of the virtualized app on the users' day-to-day activities.

Understanding the user

Compatibility testing can measure which applications can be moved to the virtual environment, and usually IT departments will focus on this. However, as part of this, user compatibility testing is overlooked. In order to truly deliver value from VDI, only those applications that will either perform better or will enhance user experience should be migrated to a virtual environment.

User compatibility testing involves a comprehension of which applications are used most by users, how they use them, from which devices, how often and for what purposes. Some applications (e.g. Microsoft Office) are more suited to VDI as they can be encapsulated easily, they are platform independent and multiple versions can run on the same box without compromising performance.

Other applications that carry a heavier data load (e.g. SAP) lose performance and are more dependent on resources outside of the discreet device accessing the data. Corresponding usage behavior of its users and an understanding of the application are essential to identifying the best candidates for VDI.

For instance, one of the behaviors that should be monitored prior to virtualization is the time required to initiate the session or begin working. If the time to log in is minimal, then it is often assumed that the user would be a good candidate for VDI or application virtualization. However, you must still look at the after state to determine if there has been an impact due to unknown issues such as network latency, shared license check-out, etc.

If this test reveals that a user will experience noticeable delays launching an app in a virtualized environment, it will likely create a negative impression and contribute to user rejection of the virtualized app. It is precisely this type of analysis and understanding that can make the difference between a successful or failed virtualization initiative.

Bite size application compatibility testing

IT departments commonly commit a one-off virtualization compatibility test before a VDI project, whereas compatibility evaluation is best conducted on an on-going basis.

For example, rather than conducting compatibility assessment for all applications used in a particular region or country (e.g. US-based apps), you're likely to ensure a higher success rate if you analyze blocks of up to a 1,000 users who are categorized by job function or department (e.g. finance). That allows IT to identify and migrate users in small groups, making the project more manageable and minimally disruptive. Getting migration right the first time must be a key goal for IT departments embarking on VDI projects.

[VDI FAQ]

The ultimate end point for the majority of enterprises today is the cloud. To successfully achieve this, a measured, best practice approach that puts users' needs at the heart of the virtualization initiative is imperative. This requires:

  • Evaluating system parameters and behaviors associated with application usage to identify the users or applications that are most suitable for migration.
  • Understanding the "before" state to compare and contrast to the "after" performance and user experience.
  • Planning the migration - the "who," "how," and "when" of the move.
  • Assessing what resources will be required - how many servers are required, what level of storage will be needed and the like.
  • Converting identified applications to the required formats ready for migration.
  • Deploying packaged applications.

Experience shows that undertaking detailed user assessment as part of the overall migration process is instrumental to the success of any VDI project. It facilitates a user-centric approach to VDI and ensures that the project truly enhances users' productivity and efficiency - in turn facilitating the achievement of enterprises' business objectives - the fundamental reason behind any IT project.

Join the discussion
Be the first to comment on this article. Our Commenting Policies