Vendor View

From Pilot to Production: 3 VDI Success Tips

VDI failures happen, but it doesn't have to happen to you. Here are three tips for success.

IT professionals jump into VDI for a number of reasons. Some get free VDI licenses as part of a server virtualization initiative. Others hear about the cost savings or reduced security risks and want to take the plunge. Whatever the reason, a successful VDI deployment can provide tremendous benefit to an IT department. Interestingly, the reality is that some VDI pilots never make it to production due to challenges that could have been avoided.

Whether the failures happen early in the evaluation process or later in the execution stage, the vast majority can be prevented with the right planning and testing, setting organizations up for success and ultimate return on investment. Here are my top three ways to ensure success of your VDI project:

1. Understand all the variables on your network
VDI allows organizations to manage, deploy and upgrade end user devices en mass, saving significant time and resources. It also enables users to log into their desktop environment from any device or location, accessing corporate documents, email, calendars, applications and more. As long as users enjoy a good experience, they will follow established protocols. But when the network becomes strained, can get frustrated, voice complaints and seek workarounds.

To prevent these issues, organizations should take time to anticipate how different users exert different workloads on the network, from the concentration of users to servers, to how similar uses cases can be grouped together. In many cases, an organization can put 100 users on a single server. However, if 90 of those users perform bandwidth-intensive activities, the other 10 may be negatively impacted. It's important to consider how users are grouped and what tasks they will be performing when projecting the network impact.

In addition, organizations with branch offices should seek VDI solutions that provide offline support that minimizes the amount of data being transferred back to the central data center, improving network performance and end user experience.

2. Anticipate costs related to future upgrades and scale
When evaluating a VDI product, it's important to consider existing infrastructure costs such as network, storage, servers and software, as well as what costs you'd incur if you decide to scale.

First, an organization needs to understand if they are purchasing an all-in-one solution or if they will have to add hardware, software or licenses. Next, they need to understand the storage implications, where surprise expenses often come up. Organizations should ask vendors how to reduce storage costs by utilizing shared storage resources, or caching different parts of the VDI storage to decrease IOPS and the amount of network bandwidth required.

Of course, VDI isn't just about today; it's also about the future. As with most software solutions, the ease of scalability is a function of the software architecture. For example, web servers are architected to scale horizontally. As web traffic increases, an organization can simply add more servers and storage capacity. While some VDI solutions follow this approach, others require the addition of multiple components, which makes scaling a challenge.

Organizations that expect to scale their VDI initiative should look for solutions that offer horizontal scalability, where increasing the number of desktops simply means increasing the number of servers and storage capacity. In this case, scaling from 200 to 200,000 users simply means adding server and storage blocks.

3. Test multiple use cases, sequenced strategically
VDI is very sensitive to the demands of different devices and use cases. Just because use case A works, it doesn't mean that use case B will also work. Too often, an organization will test just one use case, only to find themselves struggling later in the implementation as demands get more diverse and complicated.

To prevent this, organizations should test a number of use cases and personas, including what the user does, which applications they need, what times they work and which networks they access. From there, an organization can identify the most common use cases and strategically sequence them to prioritize the right resources in the right order.

In most cases, organizations should start with the easiest use cases, such as web workers, then move on to more challenging ones, such as users who access CAD functions for long hours of the day. With each scenario it's important to set user expectations, communicate changes and let them judge the success of the experience.

VDI is a powerful asset for an IT department. It can increase efficiency, improve security and enable BYOD, but it takes thoughtful planning and follow-through to ensure ongoing success. Following these three steps will ensure your pilot goes smoothly, setting the stage for a successful enterprise-wide deployment.

About the Author

Leo Reiter is Chief Technology Officer at Virtual Bridges, a provider of desktop-as-a-service private cloud solutions.


Subscribe on YouTube