Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Analyze performance with ncu. #156

Open
KuangjuX opened this issue Nov 7, 2024 · 2 comments
Open

Analyze performance with ncu. #156

KuangjuX opened this issue Nov 7, 2024 · 2 comments

Comments

@KuangjuX
Copy link
Member

KuangjuX commented Nov 7, 2024

No description provided.

@KuangjuX
Copy link
Member Author

(kM, kN, kK) (kTM, kTN, kTK)
(4096, 4096, 2048) (64, 256, 32)
cublas_time: 0.5970 ms, cutlass_time: 3.1493 ms, tiledcuda_time: 3.3075 ms

Currently, tiledcuda and cutlass have similar performance on GEMM.

@haruhi55
Copy link
Collaborator

CUTLASS can match cuBLAS's performance, the overall results are still not satisfactory.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants