eG Monitoring
 

Measures reported by VTGPUUserTest

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.

In GPU-enabled virtual environments, if users to virtual applications complain of slowness when accessing graphic applications, administrators must be able to instantly figure out what is causing the slowness - is it because adequate GPU resources are not available to the users? Or is it because of excessive utilization of GPU memory and processing resources by any of the users accessing the applications on the host? Accurate answers to these questions can help administrators determine whether/not:

  • The host is sized with sufficient GPU resources;

  • The GPUs are configured with enough graphics memory;

Measures to right-size the host and fine-tune its GPU configuration can be initiated based on the results of this analysis. This is exactly what the VTGPUUserTest helps you achieve!

To help with better utilization of resources, you can track the GPU usage rates of your instances for each user who is currently accessing the applications on the on the host. When you know the GPU usage rates, you can then perform tasks such as setting up managed instance groups that can be used to autoscale resources based on needs.

Outputs of the test : One set of results for each user who is using GPU card on the VMware Horizon View RDS server being monitored

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
No_of_sessions Indicates the number of sessions established by this user. Number  
No_of_gpu_processes Indicates the number GPU processes that are currently running in this user's session. Number  
GPU_utilization Indicates the percentage of GPU compute capability utilized by the processes running in this user's session. Percent The detailed diagnosis of this measure reveals the session ID, ID and name of process running in the session, total GPU utilization during the user session, GPU utilized for performing encoder, decoder, 3D, copy and video operations, the amount of memory that is currently in use, image path and information about the GPU card.
Encoder_utilization Indicates the percentage of GPU that is utilized for encoding processes during this user's session. 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 for any application.
Decoder_utilization Indicates the percentage of GPU that is utilized for decoding processes during this user's session. Percent
Memory_utilization Indicates the percentage of the allocated GPU memory that is currently being utilized by the processes running in this user's session. Percent A value close to 100% is a cause for concern as it indicates that the graphics memory on a GPU is almost always in use.
Memory_in_use Indicates the amount of the allocated GPU memory that is being utilized by the processes running in this user's session. MiB For better user experience with graphic applications, sufficient memory should be available to the users during sessions.
D3_utilization Indicates the percentage of GPU utilized for processing 3D frames in this users's session. Percent Compare the value of this measure across the users to identify which user is over-utilizing the GPU for processing 3D frames.
Copy_utilization Indicates the percentage of the GPU utilized for copying operations performed during this user's session. Percent Compare the value of this measure across the users to identify which user is over-utilizing the GPU for copying operations.
Video_utilization Indicates the percentage of GPU utilized for performing video decoding process during this user's session. Percent Compare the value of this measure across the users to identify which user is over-utilizing the GPU for video decoding.