Monitoring Metrics to Understand

by | May 4, 2020 | Citrix, Desktop Virtualization, Microsoft, Monitoring, Windows 10/11 | 0 comments

When you are troubleshooting or when you have a life environment. It is important that your environment is performing well. We would say “Measuring is Knowing”. Some of the most commonly used monitoring tools for VDI out there are:

  • ControlUp: Is specialized in monitoring IT Virtualization products. It is a very impressive product, designed to help the VDI SysAdmins understand their environment.
  • Uberagent: Initially created by @HelgeKlein one of the EUC community members with an impressive knowledge about anything VDI. Uber agent is a Splunk agent for end-user computing.
  • eG Innovations: A more general monitoring tool, but well known and appreciated within the EUC community.
  • Lakeside Software: Monitoring, quantifying and analyzing end-user experience. 
  • Other more general tools like Microsoft SCOM, PRTG, Zabbix, …

But what help is it to have a wonderful measuring/monitoring tool if you don’t understand what those figures actually mean. Below I list some of the more important and interesting metrics when it comes to a well performing Virtual Desktop:

User Experience

These are 2 Citrix metrics. In short:

  • ICA Latency is your network. How long does a package take to travel between your Endpoint (Citrix Workspace App) and your VDA. 
  • ICA RTT is your network + the application: The application takes it processing time to show something on the screen, this delay is added to the ICA Latency to give a number in milliseconds that a user is waiting for a return on his/her screen.

And since Windows 2019 / Windows 10 1809 there is also a Microsoft Metric:

  • User Input Delay, which is the amount of time any user input (such as mouse or keyboard usage) stays in the queue before it is picked up by a process, and the counter works in both local and remote sessions.
    • It is important to know that this will not work out of the box. You need to enable the performance counter through a registry update and a reboot.

Some very good, detailed explanations can be found:

https://www.controlup.com/controlup-automation-user-input-delay-vs-citrix-ica-round-trip-time-rtt/ 

https://docs.microsoft.com/en-us/windows-server/remote/remote-desktop-services/rds-rdsh-performance-counters 

Cognitive Load and Process Metrics

While user experience metrics are instrumental in assessing performance from a user’s perspective, we must also consider the load on the server and the efficiency of our processes. Some key metrics in these areas include:

  1. CPU Usage: This is the percentage of time the CPU is working. High CPU usage could indicate an overworked system, leading to slower response times and, consequently, a diminished user experience.
  2. Memory Usage: This metric measures how much of the system’s RAM is being utilized. If memory usage is consistently high, you might need to increase the system’s RAM to improve performance.
  3. Disk I/O: Disk Input/Output measures how much data your system reads from and writes to your storage device. High Disk I/O can indicate that your system is handling too much data at once, which might lead to slow performance.
  4. Network Latency: Network latency measures the delay that occurs in data communication over a network. High latency can result in delayed response times, affecting the user experience.

These are just a few of the metrics you can monitor to assess your VDI’s performance. However, the usefulness of these tools is contingent upon your understanding of the metrics they provide. A good grasp of these metrics enables you to make informed decisions about your infrastructure, optimize system performance, and ultimately deliver a superior user experience.

Advanced Monitoring and Predictive Analysis

With advancements in technology, more sophisticated methods of monitoring and analysis have become available. Machine learning algorithms can now analyze large amounts of data from your environment, identify patterns, and predict future behaviors. This can help in identifying potential issues before they occur, allowing you to take preemptive action and maintain a high-performing VDI environment.

  1. Session Duration: This metric measures the length of a user’s session. Abnormally short or long sessions could indicate an issue that needs addressing.
  2. Application Load Time: How quickly do applications in your environment load? Slow load times could be a sign of a performance issue and affect user experience.
  3. Active Session Load: This indicates how many active sessions are on your VDI at any given time. A sudden increase in active sessions could suggest that your infrastructure is under more strain than usual.
  4. Logon Duration: This measures the time it takes for a user to log in to the VDI. Long logon times could be a sign of a performance issue.

Performance Counters and Fine-tuning

Windows Performance Counters can be a helpful tool for monitoring your system’s performance, providing real-time data about hardware and system services. By enabling these counters, you can gain additional insights that could help optimize your environment.

However, it’s important to remember that monitoring tools and performance counters are not a cure-all solution. They are most effective when used in conjunction with other best practices, such as regularly updating and patching your system, optimizing configurations, and regularly auditing your environment.

By leveraging the right tools, understanding key metrics, and implementing sound maintenance practices, you can create a high-performing VDI environment that meets your users’ needs and supports your organization’s goals.