Song-Ling Huang

FabricPerf: Measuring NIC-less Scale-Up Network through GPU Communication Kernel Profiling

This work implements FabricPerf for the CUDA/NvLink ecosystem and validate its high precision and low overhead via extensive experiments on H100 and GB200 NVL72, and believes FabricPerf improves Scale-up network observability, laying a foundation for more complete and efficient network stacks.