GPU Monitoring

Monitor Graphic Processing Units (GPU) to ensure smooth functioning of application rendering, video playbacks, other graphics-intensive workloads.

Start 30-day free trial Try now, sign up in 30 seconds

A GPU is used for rendering images, videos, and 2D/3D animations for display, that processes large blocks of data parallelly. Install this plugin to get total visibility into the GPU resources, enable resource planning, and resolve bottlenecks before they impact end users.

Performance Metrics

  • Memory utilization
  • CPU utilization
  • Hardware metrics like temperature
  • Individual core utilization
  • Device name

Prerequisites

  • Ensure the python module 'gpustat' is installed to get the performance metrics. Execute the following command to install the module.
    pip install gpustat
  • The machine command 'gpustat' has to be installed or present in the server to get the individual core utilization metric.
  • Our Linux server monitoring agent should be installed in the server.
  • While adding a plugin, the plugin name and its folder name inside the Site24x7 agent should be identical.

Plugin Installation

Linux

  • Download and install the latest version of the Site24x7 Linux agent in the network where you plan to run the plugin. If it is installed successfully, you will see a Linux server monitor in the Site24x7 Control Panel. This confirms that the agent is able to communicate with our data center.
  • Download the gpu_monitoring.py file from our GitHub repository.
    wget https://raw.githubusercontent.com/site24x7/plugins/master/gpu_monitoring/gpu_monitoring.py
  • Create a folder with the name 'gpu_monitoring', under the Site24x7 Linux agent plugin directory '/opt/site24x7/monagent/plugins/' and place the 'gpu_monitoring.py' file under '/opt/site24x7/monagent/plugins/gpu_monitoring/'.
The agent will automatically execute the plugin within five minutes and send performance data to the Site24x7 data center.

View Data in the Site24x7 Web Client

  1. Log in to Site24x7 and go to Server > Plugins > click on the name of the plugin monitor.
  2. You will be able to view the performance metrics for the GPU resources.

Plugin Contribution

Feel free to contribute to our existing plugin and come up with suggestions or feedback on our Community.