How SAP IT saves hundreds of hours of end-user down time through Application Virtualization
In the previous posts, we have looked at how we keep our client applications “always up-to-date” and how you can develop your own Application Virtualization strategy. But how does all this fit together? How and where can IT save costs in particular? And how does Application Virtualization improve the software provisioning process to your end-users?
Let’s turn to the five fundamental phases that each managed client software package passes throughout its lifecycle in the IT environment – regardless of it being a classic (physical) or a virtual package.
We will compare two sample applications to visualize the potential savings through the use of Application Virtualization. Every standardized software installation needs some involvement from IT (whether it is a manual installation or preparing a package for automated installation) and there is typically some down time for the end-user while the software package is being installed.
Gather requirements from application owners and estimate the time it will take for your software packaging team to prepare the application for automated installation and deployment.
In order to ensure that end-users are running the latest versions of their applications, companies spend hundreds of hours per application per rollout and update. The charts below are a comparison of the amount of time needed for installing an application on the end-users’ system using three different methods: Manual Installation, Software Deployment, and Application Virtualization.
You can create these charts for every application in your environment to help you decide which deployment method to use for a particular software package.
The left chart is a detail of the first 100 installations to highlight how quickly the ROI can be achieved.
The right chart visualizes the potential for saving time and costs by using Application Virtualization (for 1000 installations) – and this is just for one package!
Application A is a fairly small and easy to package application (classic package: ~2 days / virtualized: ~1 day), while Application B is bigger and more complex (classic package: ~4 days / virtualized: ~2 days). Estimated savings compared to automated installation (per 100 installations): Application A: approx. 22 hours / Application B: approx. 70 hours.
It is important to note that although it is feasible to automate the installation of a software product (and this has been common practice in IT for the past 10-15 years), virtualized applications can be deployed even faster and more efficiently – reducing the down time per user and thus reducing the software lifecycle costs within your company.
Cost and time savings are not only achieved during the installation phase. Imagine the vendor of Application A frequently releasing updates to close security vulnerabilities. By virtualizing this application, IT can achieve these savings (see charts above) during every patch cycle!
Virtualized applications can typically be removed from the target system very easily and very quickly. Think of it as deleting a file – it is fast, reliable, and it doesn’t leave any traces behind.
The application is removed from environment and potentially replaced with a newer version of the product, initiating a new lifecycle for the new version – bringing us full circle.
Have you evaluated Application Virtualization in your environment, yet? Have you seen similar potential?
 Andreas Welsch – Applikationsvirtualisierung – Untersuchung geeigneter Einsatzbereiche als Ergänzung oder Alternative bestehender Softwareverteilungs- und -installationsverfahren sowie prototypische Umsetzung (“Application Virtualization – Evaluation of potential usage scenarios as an addition or alternative to existing software deployment and installation procedures, and prototypical implementation”) – August 2010
 Manual Installation = number of Users * (IT installation time + end-user down time)
Software Deployment = IT packaging time + number of Users * end-user down time
Application Virtualization = IT packaging time + number of Users * end-user down time