gpu-architecture

2 posts tagged with "gpu-architecture"

July 9, 2026

Redesigning GROMACS Halo Exchange: Improving Strong Scaling with GPU-initiated NVSHMEM

GPU-initiated NVSHMEM redesign of GROMACS domain decomposition halo exchange, fusing data packing and communication to eliminate CPU-GPU sync bottlenecks and improve strong scaling by up to 2x

system-optimization kernel-optimization distributed-training communication gpu-architecture

July 8, 2026

A Performance Model for Warp Specialization Kernels

A performance model using differential equations to accurately predict execution time of warp specialization kernels, enabling optimal tile configuration search via Z3 SMT solver

kernel-optimization system-optimization gpu-architecture performance-model