eG Monitoring
 

>Measures reported by VmgGPUAdapterTest

A graphics processing unit (GPU) adapter (also known as video card, graphics card, and video adapter) is a discreet card connected to one or more physical/virtual machines. The GPU adapter contains video RAM and a GPU chip and renders real-time 2D and 3D graphics applications, images, animations and video by performing rapid mathematical calculations. To perform such calculations at higher speed, the GPU adapter should be allocated with adequate memory on the virtual/physical machine. Inadequate memory allocation and inefficient utilization of the GPU adapter can affect the processing of intensive tasks related graphics on the virtual/physical machines, which in turn can adversely impact the performance of the applications operating on the VMs. Therefore, it is imperative that administrators should closely observe how well the GPU adapter uses the allocated and shared memory available to it. This way, a potential memory crunch can be proactively detected. This can be achieved with the help of the VmgGPUAdapterTest test.

The test auto-discovers the GPU adapters connected to each virtual/physical machine on the virtual/physical server and for each GPU adapter, reports the statistics related to memory utilization. This test also reveals how well the GPU is utilized for performing different tasks such as video decoding, processing 3D frames, etc. These metrics help administrators to judge whether/not adequate memory is available for use by the GPU adapters and identify the GPU adapter that is running out of memory.

Outputs of the test : One set of results for every combination of each virtual/physical machine:GPU adapter.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Total_avail_gpu Indicates the total amount of memory allocated for this GPU adapter on this VM. MB  
Used_gpu Indicates the amount of memory that is utilized by this GPU adapters. MB  
Free_gpu Indicates the amount of memory that is available for use by this GPU adapter. MB Ideally, the value of this measure should be high. A consistent decrease in the value of this measure is indicative of a steady erosion of memory available for GPU, which if left unattended, can significantly impact GPU functioning.
Usage_gpu Indicates the percentage of memory that is utilized by this GPU adapter. Percent A low value is desired for this measure. If the value of this measure grows very close to 100%, it indicates that the GPU will soon run out of memory resources. Such an occurrence will not only impact the functioning of the GPU card, but also the applications depend on it.
Total_dedicated Indicates the total amount of dedicated video memory allocated this GPU adapter. MB The dedicated video memory or Video RAM (or VRAM) is a special type of RAM that works with your computer's GPU card/adapter. The video RAM holds information that the GPU needs, including game textures and lighting effects. This allows the GPU to quickly access the info and output video to your monitor.
Used_dedicated Indicates the amount of dedicated video memory that is utilized by this GPU adapter. MB Ideally, a low value is desired for this measure. A high value for this measure indicates that the dedicated video memory is depleting rapidly. When the dedicated video memory is filled up, the GPU will rely on standard RAM for processing video frames and display textures. This in turn may seriously impact your overall video viewing experience.
Free_dedicated Indicates the amount of dedicated video memory that is available for use by this GPU adapter. MB A high value is desired for this measure. If the value of this measure is very low, it indicates that the GPU does not have adequate memory for processing video frames.
Dedicated_usage Indicates the percent of dedicated video memory that is utilized by this GPU adapter. Percent A value close to 100% is indicative of a potential video memory contention. In such a situation, use the detailed diagnosis of this measure to find out the names of top-10 applications that are over-utilizing the dedicated memory. Administrators can also find out the path to the folder where each application image is stored.
Total_sys_shared Indicates the total amount of system memory that can be utilized by this GPU adapter. MB Shared memory represents system memory that can be used by the GPU. Shared memory can be used by the CPU for processing normal system tasks needed or as “video memory” for the GPU while processing video tasks.
Used_sys_shared Indicates the amount of system memory that is utilized by this GPU adapter. MB Ideally, a low value is desired for this measure.
Free_sys_shared Indicates the amount of system memory that is available for use by this GPU adapter. MB The value of this measure should be high.
Sys_shared_usage Indicates the percentage of system memory that is utilized by this GPU adapter. Percent A value close to 100% indicates that the GPU adapter is rapidly running out of system shared memory. In such a situation, use the detailed diagnosis of this measure to find out the names of top-10 applications that are over-utilizing the shared memory. Administrators can also find out the path to the folder where each application image is stored.
Sys_video_memory Indicates the amount of system video memory that can be used by this GPU adapter. MB  
Total_commited_mem Indicates the total amount of committed memory allocated for this GPU adapter. MB Use the detailed diagnosis of the Total committed memory measure to find out the names of top-10 applications that are over-utilizing the GPU committed memory. Administrators can also find out the path to the folder where each application image is stored.
GPU_utilization Indicates the percentage of GPU resources available to this adapter that is in use currently. Percent A value close to 100% is a cause of concern which requires further investigation. Compare the value of this measure across the GPU adapters to know the GPU adapter that is being over-utilized.

The detailed diagnosis of this measure lists the names of top-10 applications in descending order based on the percentage of overall GPU utilization. Administrators can also find out the PID of application and path to a folder where the application image is stored.
D3 Indicates the percentage of GPU resources available to this GPU adapter that is utilized for processing 3D frames. Percent Compare the value of this measure across the GPU adapters to identify the GPU adapter that is very busy processing GPU-intensive 3D frames.

The detailed diagnosis of this measure lists the names of top-10 applications in terms of their 3D usage. Administrators can also find out the path to the folder where each application image is stored.
Video_encode Indicates the percentage of GPU resources on this GPU adapter that is currently utilized for performing video encoding. Percent Compare the value of this measure across the GPU adapters to identify the GPU adapter that is very busy encoding video. The detailed diagnosis of this measure lists the top-10 applications in terms of how much GPU resources they spent on video encoding. Administrators can also find out the path to the folder where each application image is stored.
Video_decode Indicates the percentage of GPU resources on this GPU adapter that is currently utilized for performing video decoding. Percent Compare the value of this measure across the GPU adapters to identify the GPU adapter that is very busy decoding video. The detailed diagnosis of this measure lists the top-10 applications in terms of how much GPU resources they spent on video decoding. Administrators can also find out the path to the folder where each application image is stored.
Copy Indicates the percentage of GPU resources on this GPU adapter that is currently utilized for copying operations. Percent Compare the value of this measure across the GPU adapters to identify the GPU adapter that is very busy performing copying operations. The detailed diagnosis of this measure lists the top-10 applications in terms of how much GPU resources they spent on performing copying tasks. Administrators can also find out the path to the folder where each application image is stored.
Video_processing Indicates the percentage of GPU resources on this GPU adapter that is currently utilized for processing video frames. Percent Compare the value of this measure across the GPU adapters to identify the GPU adapter that is very busy processing video frames. The detailed diagnosis of this measure lists the top-10 applications in terms of how much GPU resources they spent on video processing. Administrators can also find out the path to the folder where each application image is stored.