🗃️ NVIDIA
3 items
🗃️ AMD
1 item
📄️ Multi-GPU Profiling
Profile multiple GPUs and hosts simultaneously with zero extra configuration
📄️ Inference Server Metrics
Automatic application metrics extraction from AI inference servers including vLLM, SGLang, and NVIDIA Dynamo-Triton
📄️ SLURM-Aware GPU Profiling
Attribute GPU and CPU profiles and metrics to SLURM jobs, and slice by SLURM user, job name, and job ID
📄️ Metrics Reference
Every metric Zymtrace collects — host, NVIDIA GPU (including MIG), AMD GPU, AWS Neuron, and inference servers (vLLM, NVIDIA Dynamo-Triton, SGLang).