rapidsai/jupyterlab-nvdashboard
A JupyterLab extension for displaying dashboards of GPU usage.
What it solves
JupyterLab NVdashboard provides a way for developers to monitor GPU hardware metrics directly within their JupyterLab environment, eliminating the need to switch between the IDE and external monitoring tools.
How it works
It functions as a JupyterLab extension that visualizes system hardware metrics. It tracks and displays data on GPU compute utilization, memory consumption, PCIe throughput, and NVLink throughput. It also includes a "GPU accelerator" button that enables GPU-backed execution for pandas and scikit-learn code (via cuDF and cuML) without requiring code changes.
Who it’s for
Data scientists and ML engineers who use JupyterLab and rely on GPU acceleration for their computations.
Highlights
- Real-time GPU Monitoring: Visualizes compute, memory, and throughput metrics.
- Time Series Analysis: Includes a brushing feature to select and inspect specific time ranges in charts.
- Synced Tooltips: Synchronizes timestamps across multiple charts for cohesive data analysis.
- Zero-Code Acceleration: One-click activation for GPU-backed pandas and scikit-learn execution.
- Theme Integration: Automatically adapts to JupyterLab's light and dark themes.
Related
- Project
- Project
- Project
- Project
- Project