Skip to main content

Separating fact from fiction when choosing a virtual desktop

(Image credit: Image source: Shutterstock/bluebay)

When it comes to choosing a virtual desktop, it can be daunting to sort through the various options. One of the most confounding considerations is how to decide whether to go with persistent or non-persistent desktops. Do your users need full administrative control, or do they just need access to a stateless desktop? In most cases – but not all – persistent desktops are the better choice. But you may not know that, based on a number of misconceptions still floating around. Let’s set the record straight.

Myth 1: Desktop management is simpler for non-persistent desktops with layers

Ideally, for large global organizations, the operating system (OS) image would be patched at fixed monthly or quarterly intervals, and the desktop pools would have tens of thousands of desktops. Your IT team would only have to manage a few applications. If your use case is served by one golden image, and the packaging overhead is minimal, then life could be pretty simple.

In the real world, though, an organization usually has hundreds – or even thousands – of applications that require frequent updates. IT teams also usually have stringent requirements for handling critical patches. Under such circumstances, having one process for dealing with physical PCs and a second process for dealing with virtual desktops actually adds complexity. Do you want to support user-installed apps? Who do you call when an application crashes? What kind of apps cannot be packaged in a layer – what about the vendor’s guest agent itself? How many base images are required? Is application layering mandatory to reduce the number of golden images?

Isolation, layering and other application virtualization techniques are held up as the ideal way to simplify desktop management. However, you have to ask: how many consoles are required to implement non-persistent pools with application layers, Unified Endpoint Management and profiles? You’ll need to consider how many databases and load balancers are required to deploy the first desktop pool, and what changes are required when you have to deploy desktops at a new global site. How many desktops should be in the pool, and how many apps need to be packaged before a company will see real benefits? These and other questions, including those of compatibility and performance and storage overhead associated with application layering, must be answered.

It’s very hard to achieve truly stateless desktops. For most customers with limited IT staff, there are diminishing returns if every application in the company has to be converted to a new format and expensive consultants are required to implement “user layers.” Ultimately, it becomes clear that there are (too) many steps required to deploy your first non-persistent desktop.

Myth 2: Non-persistent desktops offer better security

In their purest form, non-persistent desktops remove all the user state when a user logs off. Every time a user logs in, they get a clean desktop, and the base image is patched frequently. In addition, there are no administrative privileges – the desktops are locked down – and end users have access to only a few applications. It’s true that for such deployments, non-persistent desktops are more secure. However, it’s critical to note that this 100 percent stateless desktop is only applicable to a few use cases.

Non-persistent desktops are technically stateful in nature, due to the network profiles and layering technologies. The desktops keep all the state on the network or on a writable volume. So, is IT scanning all the user profiles for downloaded content? What’s the anti-virus and anti-malware policy for such desktops? If the desktop is assembled on the fly, with “stateful” user layers, the security benefits of stateless desktops no longer apply.

Myth 3: Reduced cloud costs result from non-persistent desktops

As organizations try to save money, the standard model is to configure your virtual desktop environment with as many users as possible sharing as few resources as possible. That seems to make sense. However, the question now becomes: What can effectively be shared? Rather than a dedicated Windows 10 OS virtual machine (VM) with a certain level of CPU, memory, storage and networking speed per user, you’d share several Windows Server OS VMs with much larger amounts of those same resources for all of your users to share. How many servers will you need, and how many users can you put on each server? You’re looking at a tradeoff between user performance and overall experience – plus the cost of all those servers. Piling many users onto one server might save money in the first place, but you’ll wind up spending more money when the users complain about poor performance and latency.

So, when you include the additional software cost and complexity to manage the user profiles and app layering, you’re approaching the cost of a persistent cloud desktop per user, particularly when the users require collaboration tools, full browser-based multimedia experiences and rich applications. It will be hard to defend the position that your costs are lower than a persistent desktop.

Myth 4: Storage performance is better with non-persistent desktops

The first major issue related to VDI adoption was storage. Some vendors came up with non-persistent desktops that used linked clones with a shared base OS disk to make sure that the read performance/boot storms are addressed with heavy caching on every host in the cluster.

But things have changed a lot in the past decade. Now, every VM is guaranteed enough read/write input/output operations per second (IOPS) to handle all manner of input/output problems with modern cloud architectures. Thin or shadow clones have made linked-clones completely redundant. With all-flash arrays and widespread adoption of hyper-convergence and modern per-host caching in hypervisors, storage isn’t a bottleneck anymore. So, storage performance is no longer a valid argument.

Debunking the misconceptions

The idea of non-persistent desktops seems to make sense initially, but that’s primarily because outdated information has created a set of myths about them. Security, storage performance, management and costs are not such clear-cut issues as they once were. As it turns out, these issues have shifted since VDI was introduced such that the tables have turned. In short, persistent desktops are the better choice in most cases. Use the information above to determine what will best serve your organization’s virtual desktop needs.

Brad Peterson, VP of Marketing, Workspot