Writing advanced CUDA GEMM
How to write CUDA kernels that optimizes all compute units of GPU
Recent posts
Retrospect of ASPLOS 2024
My personal impressions from attending ASPLOS 2024 in San Diego, USA.
TCCL: Discovering Better Communication Paths for PCIe GPU Clusters
Heehoon Kim, Junyeol Ryu, and Jaejin Lee. ASPLOS, April 2024.
Retrospect of ICPADS 2023
My personal impressions from attending ICPADS 2023 in Hainan, China.