# Liu-xiandong/How\_to\_optimize\_in\_GPU

<https://repo.masters-masters.com/en/repos/Liu-xiandong/How_to_optimize_in_GPU>

[Home](https://repo.masters-masters.com/en) › [Liu-xiandong](https://repo.masters-masters.com/en/owners/Liu-xiandong) › [How\_to\_optimize\_in\_GPU](https://repo.masters-masters.com/en/repos/Liu-xiandong/How_to_optimize_in_GPU)

Liu-xiandong/How\_to\_optimize\_in\_GPU has 1,370 stars on GitHub. Its RepoMasters card: Beta tier. This is a series of GPU optimization topics. Here we will…

| Field | Value |
| --- | --- |
| Owner | [Liu-xiandong](https://repo.masters-masters.com/en/owners/Liu-xiandong) |
| Stars | 1,370 |
| Forks | 189 |
| Language | [Cuda](https://repo.masters-masters.com/en/languages/cuda) |
| License | Apache-2.0 |
| Created | 2021 |
| Topics | gpu-acceleration, elementwise, reduce, sgemm, sgemv, high-performance-computing, hpc |
| Tier | [Beta (700 to 1,999 ★)](https://repo.masters-masters.com/en/ranks/box_office) |
| View on GitHub | [Liu-xiandong/How\_to\_optimize\_in\_GPU](https://github.com/Liu-xiandong/How_to_optimize_in_GPU) |

## Repository

This is a series of GPU optimization topics. Here we will introduce how to optimize the CUDA kernel in detail. I will introduce several basic kernel optimizations, including: elementwise, reduce, sgemv, sgemm, etc. The performance of these kernels is basically at or near the theoretical limit.

The tier for repos with 700 to 1,999 ★. Published odds per card pulled: 6%. Rarity: Rare.

## Nearby in Cuda

- [mit-han-lab/torchsparse · ★ 1,477 · Cuda · Beta](https://repo.masters-masters.com/en/repos/mit-han-lab/torchsparse)
- [vlfeat/matconvnet · ★ 1,427 · Cuda · Beta](https://repo.masters-masters.com/en/repos/vlfeat/matconvnet)
- [k2-fsa/k2 · ★ 1,370 · Cuda · Beta](https://repo.masters-masters.com/en/repos/k2-fsa/k2)
- [udacity/cs344 · ★ 1,357 · Cuda · Beta](https://repo.masters-masters.com/en/repos/udacity/cs344)
- [siboehm/SGEMM\_CUDA · ★ 1,355 · Cuda · Beta](https://repo.masters-masters.com/en/repos/siboehm/SGEMM_CUDA)
- [MoonshotAI/FlashKDA · ★ 1,295 · Cuda · Beta](https://repo.masters-masters.com/en/repos/MoonshotAI/FlashKDA)

---
Data: GitHub · Illustration drawn by RepoMasters from the repo’s ID and its language color.