eG Monitoring
 

Measures reported by WVDGPUUserTest

GPU-accelerated computing is the use of a graphics processing unit (GPU) together with a CPU to accelerate scientific, analytics, engineering, consumer, and enterprise applications. GPU-accelerated computing enhances application performance by offloading compute-intensive portions of the application to the GPU, while the remainder of the code still runs on the CPU.

If a GPU-enabled session host / AVD is not sized with adequate GPU resources, then any user accessing graphic applications on that host is bound to complain of slowness. Also, if even a single user of an AVD is accessing graphic applications that are utilizing GPU memory and processing resources excessively, it could cause a GPU contention on the AVD that may adversely impact the graphic experience of other AVD users. To ensure an above-par graphic experience for all AVD users, administrators must track how much GPU resources are used up by each user, and proactively identify that user who may potentially cause a GPU crunch on a chosen session host / AVD. This can be achieved using the WVDGPUUserTest. This test auto-discovers the users who are accessing graphic applications on a chosen session host / AVD, and reports the memory and processing resources used by each user. In the process, the test points you to users who are accessing resource-hungry graphic applications on the AVD.

Outputs of the test : One set of results for each user who is accessing graphic applications on the chosen Session Host / Azure Virtual Desktop that is using GPU.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
No_of_sessions Indicates the number of open sessions for this user. Number  
No_of_gpu_processes Indicates the number of instances of graphic applications that are currently running across all sessions of this user. Number  
GPU_utilization Indicates the percentage of time for which GPU was utilized by this user's sessions. Percent A value close to 100% indicates that the a particular is hogging the GPU resources.

To zoom into the exact session where maximum GPU was consumed, and the precise graphic application that was accessed during that session, use the detailed diagnosis of this measure.
Encoder_utilization Indicates the amount of GPU that graphic applications accessed by this user are utilizing for the encoding process. Percent A value close to 100 is a cause of concern. By closely analyzing these measures, administrators can easily be alerted to situations where graphics processing is a bottleneck.
Decoder_utilization Indicates the amount of GPU that graphic applications accessed by this user are utilizing for the decoding process. Percent
Memory_utilization Indicates the percentage of time during which memory on the GPU was read from/written to by the graphic applications accessed by this user. Percent A value close to 100% is a cause for concern as it indicates that a user is almost always using the graphics memory on a GPU.
Memory_in_use Indicates the amount of graphics memory used by the graphic applications accessed by this user. MiB Compare the value of this measure across users to know which user is hogging GPU memory.