ESC

Type to search articles...

No articles found.

↑ ↓ Navigate ↵ Open
Esc Close
Blog Tags GitHub
All tags

gpu-optimization

3 posts tagged with "gpu-optimization"

July 4, 2026

PyGraph: Robust Compiler Support for CUDA Graphs in PyTorch

针对 PyTorch 中 CUDA Graph 部署困难的编译器框架。提出三项优化:CGCT(自动代码变换使 ML 程序兼容 CUDA Graph)、PI(参数间接寻址将数据拷贝转为指针拷贝,最高减少 99% 拷贝量)、SCG(基于成本效益分析的有选择部署)。在 25 个 ML 工作负载上,PyGraph 比 PyTorch2-CG 平均提速 29%,最高 3.36×,且在分布式 Tensor Parallelism 下表现更强。

compiler cuda gpu-optimization pytorch system-optimization
July 4, 2026

PyTorch 2: Faster Machine Learning Through Dynamic Python Bytecode Transformation and Graph Compilation

PyTorch 2 introduces TorchDynamo and TorchInductor for JIT graph compilation in PyTorch while retaining eager mode flexibility

compiler pytorch jit-compilation graph-capture triton gpu-optimization
June 25, 2026

PyGraph: Robust Compiler Support for CUDA Graphs in PyTorch

Compiler Framework to Maximize CUDA Graph Coverage and Benefits

pytorch cuda-graphs compiler gpu-optimization kernel-launch parameter-indirection code-transformation

Navigation

  • Work

Resources

  • Lexington Themes.

Socials

  • @Mike_Andreuzza
© 2025 MicroStudio. All rights reserved.

MicroStudio is not affiliated with Stripe, Breeew, Astro, or Tailwind Labs, nor is it endorsed or sponsored by them.