The case for VDI in 2021

Nothing new under the sun

Computing, like fashion, has seen a number of ideas that come, go and then return in a slightly different form. cloud is a prime example - being the latest twist on the idea of remote, powerful computers and local, “dumb” terminals, which can be traced back to the early days of mainframes.

Over the last few years companies have moved more and more applications and their corresponding data sets to the cloud. This pragmatic move has enabled businesses to benefit from the well-documented upsides of cloud computing: agility, location flexibility, scalability - all combined with a usage-based, OpEx financial model.

Not all applications play well with the cloud model, however, leading to frustrations for users and a potential erosion of the business case.

In this article we look at:

  • The challenges that can be caused by moving legacy, data-heavy applications to the cloud;

  • A potential mitigation through the use of Virtual Desktop Infrastructure (VDI); and

  • How to make the costs add up.

Distant clouds

For a particular set of users, the move to the cloud has introduced a new problem – they are now remote from their data. Anyone using modern business applications is impacted very little as these applications are well suited to the cloud environment – running queries in databases and presenting the results to users.

Older applications, however, do not fare so well. They fetch data piece-meal from databases, or even large datafiles, and then process that data locally. This didn’t cause performance issues when the data was stored just milliseconds away on a local network – but becomes more of a problem when the data could be hundreds or thousands of miles away. Indeed, displaying a single set of results may require many round trips from the user to the data and back again – drastically degrading the performance – leading to increasing user frustration and affecting business efficiency.

In the short term, to solve this problem there seem to be two solutions: move the user closer to the data or move the data closer to user.

The virtual user (or desktop)

Virtual desktops allow IT and infrastructure teams to put the “compute” part of the user’s desktop closer to the data. This means that the majority of the work happens in the cloud – with only the users “screen” sent back to them. This approach massively reduces the time taken for each request of data to go from the user’s virtual desktop to the cloud-based data stores. This architecture can be optimised by putting the user’s desktop compute in the same cloud region – right next to the data.

Balancing the cost

Moving a user’s desktop to the cloud and running it there incurs cost. Where all requirements for heavy local processing can be removed, the cost of providing cloud-based VDI can offset by reducing the spec of the standard end-user device. In enterprise office environments where staff have traditionally worked from a fixed desk, a “thin terminal” such as those made by Wyse (now owned by Dell) could be provided.  These devices are not portable, and post-COVID it may be preferable to use an alternative such as a low-powered laptop such as a Chromebook.

There are risks when pursuing a cloud VDI solution.  Some organisations continue to provide users with full-featured Windows 10 laptops and end up incurring not only the cost of maintaining and supporting a Windows laptop but the cost of the virtual desktop infrastructure in the cloud as well. This can significantly erode the financial case if you’re not careful – thus ending with higher overall infrastructure costs.

Recap - avoid the double bubble!

Using virtual desktop technology can materially help your business if you have moved your data to the cloud and are running old applications – but be careful not to erode the cost advantage of using cloud-based desktops, by supplying an on-premise one as well.

Previous
Previous

AI: The only way is ethics

Next
Next

Virtualisation in the cloud era