login vsi company logo login vsi company logo 250x40
header-05.jpg
  • Home
  • Blog
  • Beyond VDI and XenApp Benchmarking

Beyond VDI and XenApp Benchmarking

Beyond VDI and XenApp Benchmarking

I’d like to take a moment to look beyond the original use case—benchmarking—that Login VSI creators (shout out to @MarkPlettenberg) envisioned. The purpose is to help our existing Login VSI customers get even more out of the product. For enterprises and governments not yet using Login VSI, I want to lay out the full value proposition so that you have more resources for a purchase decision.

First, an obvious statement: virtualized desktop environments are complex. With users sharing centralized servers, storage, software and even GPUs, the IT department faces added difficulty offering the users the same performance they grew accustomed to on their physical devices. Over time performance can dwindle with no traceable cause. What’s more, application updates affect performance in unpredictable ways unique to your specific environment.

beyond vdi and xenapp benchmarking henk hofs and mark plettenberg

Login VSI: You’ve come so far. The original Login VSI creators Henk Hofs and Mark Plettenberg in 2008

Testing in transformation projects

Using Login VSI for greenfield or transformation projects is already a given when end user experience is a priority. Most IT departments with XenApp or moving into VDI realize performance benchmarking is a necessary step for pilots and POCs. With so many technology and service options to consider when undertaking a new or expanding an existing VDI or XenApp project, capacity planning and load testing are essential to moving confidently into production. Testing should always be performed at scale with synthetic users, before real users are live and in production.

At this point you will want to validate performance: How many users can my infrastructure support before performance suffers?

Testing in production

Once in production, use Login VSI for change impact analysis. Successful centralized desktop environments are managed and maintained with a disciplined approach to release and change management that includes (automated) performance tests.

Every update, upgrade or other change to your environment can have a negative impact on end user productivity. For example, what may appear to be a minor Microsoft update may have a serious impact on performance in a virtualized desktop environment. Many IT departments take it for granted that new software versions are controlled by vendors for a negative impact. In fact, each update will have a unique impact depending on the specifics of the environment.

In this phase of your desktop virtualization deployment, you will use Login VSI to predict: What is the performance impact of necessary updates and upgrades?

Managing your VDI environment

Earlier we talked about using Login VSI for benchmarking, load testing, and capacity planning. Then we talked about using Login VSI for change impact analysis. In production, Login VSI is now a resource to manage your VDI or XenApp environment: How do I stay ahead of support tickets from end users and understand performance from the end users’ perspective? On the road to proactive performance management, you will want to use Login PI. Only by deploying a synthetic user can you truly understand what your production end users are experiencing. Monitoring products will tell you what is happening on your servers and if CPU is maxed out, but they won’t alert you to what is going on with end users.

Finally, test at scale

Whether in transformation or production phase, @MarkPlettenberg says: “Testing does not scale linearly! Test with the total number the intended users!” and so I repeat that mantra. The main pitfall VDI or XenApp performance testing is to assume that performance scales linearly with the number of servers. Data from limited tests cannot be accurately extrapolated to the full user base. There are many examples of hosted desktop infrastructure tests sample of real users that performed well initially, but later on experienced fundamental performance and capacity issues when production was scaled out. A complete environment responds fundamentally differently than when testing a subset of users.

I hope you found this helpful. The white paper Avoid VDI Performance Problems gives more detail.

About the author

Tags: Login VSI, Login PI

What our customers are saying

Michael Bernas - Workers' Compensation Board Alberta

“We've used Login VSI for the past 3 years as part of our Disaster Recovery Exercise. It's not only helped us in validating the functionality of our Citrix XenApp farm, but it's also been an invaluable tool in proving the number of concurrent connections into the environments. It’s allowed us to take our theoretical numbers and put actual real life data behind them.”

Michael Bernas, Workers' Compensation Board of Alberta


Tristan Todd - VMware

“Login VSI is my favorite test orchestration and simulation tool for large-scale reference architecture testing. Not only do we use it in our testing, but we work with partners who use Login VSI exclusively, with the idea that you create the most realistic simulated workload at scale, to really test infrastructure and software capabilities. What’s more, getting the results and be able to measure yourself across platforms and across vendor solutions is just fantastic and is unequalled in the industry.”

Tristan Todd, EUC Architect at VMware


Michael Mainusch - Dachser

“With Login VSI, we can simulate the effects and compatibilities of applications and patches, as well as scaling in a virtualized terminal server environment, with minimal expenditure of time. Due to the adaptability of Login VSI to individual customer environments, we can simulate the real production environment, using the achieved measurement results and assertions.”

Michael Mainusch, Team Leader Application Services at Dachser

Login VSI, Inc.

300 Tradecenter

Suite 3460
Woburn, MA 01801

Phone: +1 844 828 3693

Login VSI B.V.

De Entree 85
1101 BH Amsterdam
The Netherlands

Phone: +31 20 705 1200