skip to main content
Posted on by Phil Spitze in VDI, Student Experience, Application Virtualization

Examining the costs of delivering software apps in Higher Education

One of the unique challenges in managing computing services in a university environment is how to deliver applications to students in a cost-effective manner while ensuring a good user experience.

The layers of this challenge are many:

  • Computing labs
  • Library spaces
  • Teaching podiums
  • Student-owned devices
  • Supporting alternative OSs like macOS and Linux

A short history of application delivery...

Over the years there have been many products and solutions attempting to solve this challenge. Some work better than others. Some fit only certain use cases, specifically in the corporate market. From my time working in the IT department at a Le Moyne College, I soon came to realize that many of the solutions out there don’t fit at all into the complex world of higher ed computing.

Among these software delivery solutions are:

  • Building out thick/fat images for physical workstations
  • Implementing VDI
  • Deploying App Remoting (RDS) using products from Microsoft and others 

While each of these approaches have some benefits, the one thing they all tend to have in common is high backend cost and/or lots of admin time and attention by IT staff.


Examining costs of app delivery

Let’s look at the three most common approaches one at a time and some of the associated cost estimates:

1st - FAT Deployment

Maybe the oldest method around, is the building and deployment of fat images to physical computers. Traditionally this takes place each summer, and depending on the number of managed endpoints, can take 1 to 3 (or more) employees about 8 weeks to complete.

Based on public salary data, your staff are probably in the $38,000 to $42,000 salary bracket, or roughly $80 per hour. For 8 weeks of work, re-imaging the desktops each summer costs about $22,000 in time, per individual. Of course, there are automation solutions that can help with this manual task, but the administration of such systems often requires a more highly qualified employee, so realistically the savings in time is a wash against a higher salary.

Read more: Send Ghost to the Grave >

FAT Deployment of Software using Ghost Imaging

2nd - App Remoting / RDS

Probably the next oldest method; app remoting or remote desktop solutions. Here, applications are locally installed on big backend servers with a broker or session manager framework installed. Users connect and consume the apps via a streaming protocol.

So first we’ll need some backend servers. A ballpark estimate from Dell.com puts each server at between $15,000 and $25,000 (depending on hardware options, licensing, and support requirements). Rule of thumb is that one server can handle about 30 concurrent sessions, though less if the apps have higher computing and GPU requirements (like, for example, SPSS, SAS and ArcGIS).

Let’s say 200 concurrent sessions will be needed, which results in a cost range of $105,000 to $175,000. But, because app remoting is a complex platform, you’ll need at least one dedicated system admin that will cost between $58,000 and $70,000 per year when we look at the public data.

Read more: Drop the Heavy Lift of App Remoting (RDS) >

Using app remoting and RDS to deploy applications

3rd - VDI / Desktop Virtualization

Finally, we look at VDI (Virtual Desktop Infrastructure). This is almost like the next-generation of app remoting, but instead our users are consuming an entire desktop instead of just individual apps. Here, all the apps will have been installed or delivered virtually through the VDI platform.

Now this is where things get (very) expensive. If we use the same 200 concurrent session number from before, the backend investment cost for a VDI deployment is around $800,000 – that covers the costs of fast storage, networking, computing and licensing. With this option, you’re likely going to need a pair of system admins to look after all the components, so that will cost another $116,000 to $140,000 depending upon qualifications and experience.

In a nutshell, for a small investment, Cloudpaging has enabled us to finally enjoy many, if not all, of the benefits originally sold to us regarding VDI.

Le Moyne University, Syracuse NY 

Read more: How to fix VDI to improve application delivery >

Using VDI and desktop virtualization to deploy applications

Is there a better way to deliver applications for less cost?

Well, I think so...

AppsAnywhere takes away all the cost of the backend hardware investment, and is so lightweight from an administration standpoint you can simply retrain the staff you already have. For starters, it only needs 6-8 basic VMs to support a medium-sized campus of 5000 students.

Delivering higher education applications with AppsAnywhere, the university app store

AppsAnywhere enables you to deliver 100% of Windows apps (and we mean, 100%!) to physical endpoints without having to image them every year. If you already have VDI or an App Remoting/RDS solution, it can integrate seamlessly with that, allowing you to save money by 'right-sizing' those environments. And by integrating with other software delivery technologies, we can also deliver any app to other operating systems, including macOS and Chromebooks.

You can even start to support BYOD and meet student expectations of having access to any app, anywhere on and off campus, and at any time. The best part? You can be up and running for roughly the cost of a single system admin. A far better use of departmental funds, and a guaranteed way to create an awesome student experience, whilst meeting your strategic IT department goals.

Our customers: See how other universities are using AppsAnywhere >


Try it now

The list of features and benefits goes on, but it’s best to see it in action!

Try AppsAnywhere - the university app store and virtualization solution

Try AppsAnywhere now on our trial site

Read more about delivering software apps in higher ed

Our awesome customers...

Comments 0

Leave a comment

We are always excited to hear your views and opinions on virtualization or big data at your university or college.