Workload Analyzer 📊
A simple tool for monitoring and analyzing GPU and system resource usage during AI/ML workloads.
🌟 Features
- 📈 Real-time monitoring of GPU utilization, memory usage, and system resources
- 🔍 Detailed visualization of resource usage over time
- 💡 Intelligent recommendations for workload optimization
- 🚀 Easy-to-use CLI interface
- 📝 Comprehensive statistics export
🔧 Installation
Using pip
pip install workload-analyzer
Development setup
# Clone the repository
git clone https://github.com/yourusername/workload-analyzer
cd workload-analyzer
# Install using uv with development dependencies
uv sync --dev
📋 Requirements
- Python 3.12+
- NVIDIA GPU with nvidia-smi (for GPU monitoring)
🚀 Quick Start
# Run a command with default settings
workload-analyzer "python train_model.py"
# Specify timeout and polling interval
workload-analyzer "python train_model.py" --timeout 300 --interval 5
📊 Output
The tool generates:
-
Statistics file: JSON format data with all recorded measurements
-
Visualizations:
- GPU memory usage over time
- System memory consumption
- CPU and disk utilization
- Network usage
- Process memory statistics
-
Optimization recommendations based on resource utilization patterns:
- GPU memory sizing recommendations
- Compute utilization insights
- Memory bandwidth analysis
- System resource optimization tips
All outputs are saved to workload_results/ by default (configurable with --output-dir).
🛠️ Configuration options
--timeout Time to monitor in seconds (default: 120)
--interval Polling interval in seconds (default: 3)
--recommendations Enable workload optimization recommendations (default: True)
--output-dir Directory to save outputs (default: workload_results/)
--verbose Enable verbose logging (default: True)
--version Print version information
❕ License
Package is licensed under Apache 2.0 license. Free to use as you like, but a cite of the package is welcome:
@misc{skafte_workload_analyzer,
author = {Nicki Skafte Detlefsen},
title = {Workload-Analyzer},
howpublished = {\url{https://github.com/SkafteNicki/workload_analyzer}},
year = {2025}
}
Metadata
Release files for workload-analyzer 0.0.7
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| workload_analyzer-0.0.7.tar.gz | 95.2 kB | Without provenance | Details for workload_analyzer-0.0.7.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| workload_analyzer-0.0.7-py3-none-any.whl | Python 3 | none | any | Details for workload_analyzer-0.0.7-py3-none-any.whl |
Total release size: 112.7 kB
Release files / workload_analyzer-0.0.7.tar.gz
| Download URL | workload_analyzer-0.0.7.tar.gz |
|---|---|
| Size | 95.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
5f031da5b17fea26feb373c2d554d8e30d68ac554d35847777e6c7212f946ccb
|
|
BLAKE2b-256 checksum How to use checksums |
d61f5d05e93549f9cde061faf12fa2019bf5a0f8ff3406c5f4da0e0a3834c56c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
uv/0.7.10
|
Release files / workload_analyzer-0.0.7-py3-none-any.whl
| Download URL | workload_analyzer-0.0.7-py3-none-any.whl |
|---|---|
| Size | 17.6 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a4ee3d7db281e6262d77a3730d603b5e32deec3d0bb4f70420fd0f7efa97bea0
|
|
BLAKE2b-256 checksum How to use checksums |
f70d1129e21dee017e4836ddedd73cbb7f4396e6f54692abe7a64ab4445b9c8a
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
uv/0.7.10
|
Security for workload-analyzer 0.0.7
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.