VDI Monitoring: What You Don’t Know You’re Missing

How do you know that what you understand to be all-encompassing isn’t actually quite as extensive as you might have thought? Surely when you buy an all-inclusive package, you can feel safe in the knowledge that the words all-inclusive are a dead giveaway. Apparently this is not necessarily true, as while nothing is completely perfect, so this is as true of VDI monitoring as of anything else.

Through a Negative Filter

The management tools provided by the principal vendors are undoubtedly good VDI monitoring solutions — they offer an ample, all-inclusive package which is globally distributed, fully supported and more importantly, free of charge – which in and of itself is enough to turn the head of even the least savvy CFO. So what is it that isn’t there? While looking through a negative filter is a more difficult perspective to adopt, it’s invariably worth the effort.

The Detectability of Visibility, or How Not to be Seen (TM Monty Python)

How would you define “proper visibility”? Surely it should provide a high level overall view of the system performance, its users and the infrastructure. Alas, this is not always entirely true of the leading VDI monitoring solutions, which lack some if not all real-time session and process performance data, and offer no comprehensive view of the host (hypervisor) layer, or the guest OS. This omission may not be immediately clear, but if something untoward is going on in the hypervisor layer, and you lack visibility of the full stack, it’s like smelling the smoke but being unable to see the fire; a situation which is never going to end well, and one which would likely result in significant performance issues, the overhead of which being costly and time-consuming.

ControlUp offers highly effective insight into the end user experience, with a comprehensive “dashboard” overview of all current sessions and the resource consumption thereof, indicating where hardware resources are being used most heavily, and making it quick and simple to spot potential problems.

ControlUp Host View: showing host metrics

How Do You Manage?

While nominal management tools are provided free of charge with most VDI solutions, in reality they often lack any tangible troubleshooting or maintenance functionalities, meaning that any time a problem occurs, there needs to be a prolonged detection phase followed by careful analysis and deduction, all of which, when performed on a manual level, is time-consuming, clumsy and costly.

For example, in a situation where an application has ceased to function properly on three of six existing virtual servers, the system administrator must manually log on to both a working and non-working server, and then look, compare and try to ascertain where there are differences and problems, and thereby deduce what the reason is for the application to have ceased.

ControlUp Process View: showing specific metrics, drilled-down to the process level

ControlUp offers an effective and wide-ranging troubleshooting mechanism that enables you to obtain the essential data as to what and where the problem exists, without having to touch any of the manual endpoints, resulting in little-to-no disturbance to end-users, and reduced time-wastage and costs. Also, where possible, this feature offers options and suggestions for how to fix the issue.

It’s Metrics, Baby!

While user metrics are gathered as a matter of course in any VDI monitoring solution, it is how they can be utilized that makes them worthwhile. If your monitoring solution only gathers rudimentary particulars with no real degree of precision, how useful can these metrics really be? Take logon duration as an example. The ability to accurately and precisely monitor logon duration times is critical to ensure a good end user experience. A specific analysis is required, including defined logon times, and various phases of logon execution, where each phase has its own particular set of metrics.

ControlUp provides accurate built-in metrics, which breaks down the total logon duration into four fundamental elements.

  1. Profile Load Time
  2. Group Policy Load Time
  3. Desktop Load Time – the time it takes for an explorer shell to initialize.
  4. Logon Duration Other – the time that it takes for the other components to load.

ControlUp Computer View: showing the range of metrics from the host and the guest OS

ControlUp listens carefully to specific shell events, so that it can obtain accurate measures of each of these elements. Once these events are finished, it then knows that the logon process is complete.

In Conclusion…

It is widely accepted in the market and the industry as a whole, that Citrix and VMware are the leading manufacturers of VDI solutions – and there is no challenge to this understanding. Conversely, it’s clear that there are niches within their offerings that need to be plugged in order to offer system administrators the complete and truly all-encompassing package that they deserve in order to perform their duties effectively; and that ControlUp is the ideal complement to your VDI solution.