« All posts

Dgxtop: Rust-Based Monitor for NVIDIA DGX Systems

Dgxtop offers real-time GPU and system monitoring for NVIDIA DGX systems.

Dgxtop is a high-performance system monitoring tool designed for NVIDIA DGX infrastructure. It provides real-time monitoring of GPU, CPU, memory, and disk status, aiding engineers in effectively managing system resources. Developed in Rust, it offers faster and more reliable data through direct NVML access.

Equipped with features like multi-GPU monitoring, NVLink topology, and PCIe efficiency, users can manage GPU processes, monitor system resources, and conduct detailed analyses through an interactive terminal interface.