Skip to content
#

cray

Here are 12 public repositories matching this topic...

ARCHER2 上的 PETSc MPI+OpenMP 混合并行布局研究:最大 4096 核、408 个通过审计的 runs,3D 固定规模下比 flat MPI 快 3.81×,并用 Linaro MAP profiling 解释原因。

  • Updated Aug 17, 2026
  • Python

Add this topic to your repo

To associate your repository with the cray topic, visit your repo's landing page and select "manage topics."

Learn more