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

6000 Ada results #5

Open
alexweisberg opened this issue Apr 22, 2023 · 0 comments
Open

6000 Ada results #5

alexweisberg opened this issue Apr 22, 2023 · 0 comments

Comments

@alexweisberg
Copy link

Hi,

I just got guppy up and running on a machine with a 6000 Ada GPU. Here are the results of running it:

gpu,gpu_memory,gpu_memory_used,gpu_driver_version,cuda_version,guppy_version,guppy_model,chunk_size,chunks_per_runner,num_basecallers,runners_per_device,samples_per_second
NVIDIA RTX 6000 Ada Generation,49140,7043.38362,525.85.05,12.0,6.3.9+799d5bd,fast,2000,160,,,8.14516e+07
NVIDIA RTX 6000 Ada Generation,49140,7207.21638,525.85.05,12.0,6.3.9+799d5bd,fast,2000,256,,,8.20092e+07
NVIDIA RTX 6000 Ada Generation,49140,6715.81638,525.85.05,12.0,6.3.9+799d5bd,fast,2000,512,,,8.07708e+07
NVIDIA RTX 6000 Ada Generation,49140,7207.21638,525.85.05,12.0,6.3.9+799d5bd,fast,2000,786,,,8.07198e+07
NVIDIA RTX 6000 Ada Generation,49140,7371.00000,525.85.05,12.0,6.3.9+799d5bd,fast,2000,1024,,,8.03532e+07
NVIDIA RTX 6000 Ada Generation,49140,0,525.85.05,12.0,6.3.9+799d5bd,fast,2000,2056,,,8.28481e+07
NVIDIA RTX 6000 Ada Generation,49140,3276.001638,525.85.05,12.0,6.3.9+799d5bd,hac,2000,160,,,1.51165e+07
NVIDIA RTX 6000 Ada Generation,49140,4968.59454,525.85.05,12.0,6.3.9+799d5bd,hac,2000,256,,,1.48007e+07
NVIDIA RTX 6000 Ada Generation,49140,0,525.85.05,12.0,6.3.9+799d5bd,hac,2000,512,,,1.45522e+07
NVIDIA RTX 6000 Ada Generation,49140,0,525.85.05,12.0,6.3.9+799d5bd,hac,2000,786,,,1.4388e+07
NVIDIA RTX 6000 Ada Generation,49140,5842.20546,525.85.05,12.0,6.3.9+799d5bd,hac,2000,1024,,,1.46034e+07
NVIDIA RTX 6000 Ada Generation,49140,2893.800546,525.85.05,12.0,6.3.9+799d5bd,hac,2000,2056,,,1.44155e+07
NVIDIA RTX 6000 Ada Generation,49140,14716.15236,525.85.05,12.0,6.3.9+799d5bd,sup,2000,160,,,6.36116e+06
NVIDIA RTX 6000 Ada Generation,49140,0,525.85.05,12.0,6.3.9+799d5bd,sup,2000,256,,,6.36792e+06
NVIDIA RTX 6000 Ada Generation,49140,13267.80000,525.85.05,12.0,6.3.9+799d5bd,sup,2000,512,,,6.35849e+06
NVIDIA RTX 6000 Ada Generation,49140,0,525.85.05,12.0,6.3.9+799d5bd,sup,2000,786,,,6.35824e+06
NVIDIA RTX 6000 Ada Generation,49140,12983.32854,525.85.05,12.0,6.3.9+799d5bd,sup,2000,1024,,,6.34754e+06
NVIDIA RTX 6000 Ada Generation,49140,13190.20794,525.85.05,12.0,6.3.9+799d5bd,sup,2000,2056,,,6.3568e+06

I'm a little surprised some of the samples_per_second were lower than the 4090 results in the other open issue, but it still looks really good.

A few of the runs (not always the highest chunks per runner) gave weird output, like -nan% GPU memory usage, but they still produced output.

Any suggestions on where to tweak settings further? Thanks for developing this!

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

1 participant