google-deepmind / google-deepmind/alphafold
optimize alphafold2 speed on different GPUs
Open
- Dominant language
- Python
- Stars
- 14.9k
- Forks
- 2.9k
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I have run alphafold2 on two machines with A800 and V100, both machines has NVME ssd. Alphafold2 on A800 is 3-5 times slower than V100, which is not expected as A800 is a newer GPU with higher computational power. I wonder there are any configurations of alphafold2 to maximize the use of different GPUs?
Contributor guide
Assessment
This issue has not been assessed yet.