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

[Feature request] GPU acceleration? #505

Open
calebwin opened this issue Mar 30, 2020 · 2 comments
Open

[Feature request] GPU acceleration? #505

calebwin opened this issue Mar 30, 2020 · 2 comments
Labels
enhancement New feature or request

Comments

@calebwin
Copy link

Feel free to ignore this - this is low priority and not really a feature request.

I was wondering - could augur benefit from GPU acceleration? I was looking in the distance module and was wondering if some of the "hot" code in there and maybe in other places might benefit from moving data to the GPU and doing computation there.

I have some GPGPU experience so I might be able to help with this. But it would be great to know if the developers have a sense of whether or not Augur could benefit from off-loading parts of it to a GPU.

@calebwin calebwin added the enhancement New feature or request label Mar 30, 2020
@tolot27
Copy link
Contributor

tolot27 commented Mar 31, 2020

I've observed that mafft takes longer and longer with increasing number of sequences, even if I add only new sequences. I traced it to disttbfast, a component of mafft, which uses just one thread during the later stage of the alignment. Maybe only Kazutaka Katoh, the author of mafft can solve this. Or we could try G-large-INS-1.

@calebwin
Copy link
Author

calebwin commented Apr 1, 2020

Interesting. I'll definitely look into this.

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

No branches or pull requests

2 participants