ARGUS lets 10k‑GPU clusters trace performance with <2% overhead
ARGUS provides always‑on, fine‑grained tracing for clusters over 10,000 GPUs while keeping overhead under 2%. It compresses raw kernel events 3,700× and automatically isolates anomalies such as straggling ranks or kernel stalls. Deployed for six months, it has already helped diagnose compute, communication, and pipeline issues in production LLM training.