-
Notifications
You must be signed in to change notification settings - Fork 2.2k
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
Add CUDA Allgather implementation to support tensor fusion #1480
Conversation
Very nice! Adding @abditag2 who worked on the original Allgather tensor fusion implementation. |
I am also preparing a pull request to add "Reducescatter" as a first-class operation type. There's a paper about a memory optimization for distributed training. Instead of doing 'allreduce -> optimizer' they do 'reducescatter -> optimizer -> allgather'. But to do that, I need reducescatter in Horovod :) |
1dbb0db
to
19557a9
Compare
Signed-off-by: Jesse Benson (AI) <jesseb@microsoft.com>
Signed-off-by: Jesse Benson (AI) <jesseb@microsoft.com>
19557a9
to
6caabb7
Compare
I rebased on the latest master commit, and added the CUDAOpContext class for |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM! Nice work!
Allgather with CUDA tensors does not always work with tensor fusion/CUDA.