SE Articles From the Field
Ensuring cloud users are having a good experience and identify issues that may be affecting the user experience is always critical. Stratusphere can easily be configured to show the proper metrics for users running cloud based virtual desktops.
This Document is broken into 3 different sections.
- Login Experience
- User Machine Health
- Remote Connectivity Health
- Application Health
Note: Most of the metrics talked about are in this document.
1. Login Experiance
Login Health is defined as:
- Login Under XX Seconds
- Domain Controller Response time under 500ms
- Less that 2 Abnormal Event on Average over a week.
How to Navigate to Login Health:
Detailed Login Breakdown for a user:
There is an animated GIF that shows how to navigate to the Login Details
2. User Machine Health
User Machine Health is defined as:
- CPU Less Than 80% Average per hour
- CPU Queue less than 2/CPU Core
- Memory Less Than 85% Average per hour
- Disk Response Time Less than 2ms
- Disk Queue Less than 0.01 Average per hour
- Network Latency
This is the latency from the User Machine to the Stratusphere Systems
This will vary per datacenter. Need to define numbers and document them here
How to Navigate to User Machine Health:
Note: Please refer to the SpotCheck documents listed above for more information on these metrics.
3. Remote Connectivity Health
Remote Connectivity Health is defined as:
- Session Latency: This will vary based on each user location and Datacenter Loss - Typically want to see less than 1% *
- Retransmit: Typically want to see less than 1% *
- Monitor Audio: This could indicate that Audio Offload is not working
- Monitor USB Traffic: Security and Functionality
* - This is typically on the end user side and not in the Azure Datacenters
How to Navigate to Remote Connectivity Health:
Set the Table to Protocol "RFX”. This is the protocol that a user uses to connect to Azure Virtual Desktops.
4. Application Health
Application Health definition will be different based on the application but below is some general guidelines:
- CPU & Memory Usage:
We want to compare the CPU and Memory Usage to other users to form a baseline for each version of the application. If we find that some users are using a high amount of CPU/Memory compared to the general population then there may be an issue with that application.
How to Navigate to Remote Connectivity Health:
(Use the “Process Name” Tab not the “Application” Tab for higher granularity.
Pick the Process Name and Drill Down
Change the Resolution to “Process Name”
This will provide the “Base” Line for all users. Look at the “Per App Average” Columns
For per user and per machine.
Turn on the Basis “Process Owner Name” and “Machines”
This will display users with higher than normal CPU/Memory Used