Skip to main content

A simple GPU monitoring tool.

Project description

GPU Monitor (gpumon)

gpumon is a real-time GPU monitoring tool designed to display various metrics for NVIDIA GPUs, including temperature, fan speed, memory usage, load, and power consumption. It provides color-coded output for easy identification of critical values and can operate in both continuous and one-time monitoring modes.

Features

  • Real-time Monitoring: Continuously monitors NVIDIA GPU statistics.
  • Detailed Metrics: Displays temperature, fan speed, memory usage, GPU load, and power consumption.
  • Color-Coded Output: Highlights different levels of each metric with distinct colors for easy visualization.
  • Customizable: Users can adjust the bar length and refresh rate.
  • Single or Continuous Mode: Option to display GPU status once or continuously update.

Installation

Prerequisites

  • Python 3.x
  • NVIDIA drivers
  • nvidia-smi command-line utility

Install using pip

To install gpumon, use the following command:

pip install gpumon

Usage

Run the tool using the command line:

Continuous Monitoring (default mode)

gpumon

This command will start gpumon in continuous monitoring mode, refreshing the display every 0.5 seconds.

One-Time Status Display

To display the GPU status once and then exit:

gpumon --continuous False

Custom Bar Length and Refresh Rate

You can customize the appearance and behavior of the monitoring display:

gpumon --bar-length 5 --refresh-rate 1.0
  • --bar-length: Sets the length of the progress bars (default is 5).
  • --refresh-rate: Sets the refresh rate in seconds (default is 0.5 seconds).

Command-Line Arguments

  • --bar-length: The length of the progress bars. This controls how much detail is displayed in the bar representation.
  • --refresh-rate: The rate in seconds at which the display refreshes.
  • --continuous: Set to False for a one-time status display, otherwise the tool runs continuously (default is True).

Example Output

[0]-NVIDIA A100              Fan:  70% Power: 250.00 W Temp:  72C [███░░] Load:  90% [███░░] Mem:  80000/80000 MB [██████] Freq: 1500/2100 MHz [███░░]
[1]-NVIDIA A100              Fan:  75% Power: 245.00 W Temp:  73C [███░░] Load:  85% [███░░] Mem:  78000/80000 MB [██████] Freq: 1550/2100 MHz [███░░]
[2]-NVIDIA A100              Fan:  65% Power: 260.00 W Temp:  70C [███░░] Load:  80% [███░░] Mem:  77000/80000 MB [██████] Freq: 1600/2100 MHz [███░░]
[3]-NVIDIA A100              Fan:  68% Power: 255.00 W Temp:  71C [███░░] Load:  75% [███░░] Mem:  76000/80000 MB [██████] Freq: 1650/2100 MHz [███░░]
[4]-NVIDIA A100              Fan:  72% Power: 240.00 W Temp:  74C [███░░] Load:  95% [███░░] Mem:  80000/80000 MB [██████] Freq: 1700/2100 MHz [███░░]
[5]-NVIDIA A100              Fan:  60% Power: 250.00 W Temp:  72C [███░░] Load:  85% [███░░] Mem:  79000/80000 MB [██████] Freq: 1500/2100 MHz [███░░]
[6]-NVIDIA A100              Fan:  70% Power: 250.00 W Temp:  72C [███░░] Load:  90% [███░░] Mem:  80000/80000 MB [██████] Freq: 1500/2100 MHz [███░░]
[7]-NVIDIA A100              Fan:  70% Power: 250.00 W Temp:  72C [███░░] Load:  90% [███░░] Mem:  80000/80000 MB [██████] Freq: 1500/2100 MHz [███░░]

License

This project is licensed under the MIT License - see the LICENSE file for details.


Notes

  • Ensure that the nvidia-smi tool is available on your system, as gpumon relies on it to gather GPU metrics.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

gpumon-0.0.7.tar.gz (4.6 kB view details)

Uploaded Source

Built Distribution

gpumon-0.0.7-py3-none-any.whl (4.9 kB view details)

Uploaded Python 3

File details

Details for the file gpumon-0.0.7.tar.gz.

File metadata

  • Download URL: gpumon-0.0.7.tar.gz
  • Upload date:
  • Size: 4.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.8.19

File hashes

Hashes for gpumon-0.0.7.tar.gz
Algorithm Hash digest
SHA256 591318ad4636d164499f1d37059437dd396fbff571bddfe648e80d8ebd61e6e4
MD5 f68057eb1eede5d7bba95f9dead36765
BLAKE2b-256 db17083b064b05ec2200bc877a91472b199c8d82806d658de7c1c50109f6a300

See more details on using hashes here.

File details

Details for the file gpumon-0.0.7-py3-none-any.whl.

File metadata

  • Download URL: gpumon-0.0.7-py3-none-any.whl
  • Upload date:
  • Size: 4.9 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.1.1 CPython/3.8.19

File hashes

Hashes for gpumon-0.0.7-py3-none-any.whl
Algorithm Hash digest
SHA256 6249872febb86a2e6c5830aa591d6439a9b8d3cbd0b506a33310249e78b08d41
MD5 9dc3da60112e9494ab5a49ba6c1ebeb4
BLAKE2b-256 8e38c3fe0c4d683789a3bdd7aade4d5ea2f67a4c8143f928496f9e9706c951a5

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page