On 4 May 2019, at 19:16, Dev vasu <[log in to unmask]> wrote:
Dear moises,
I am using a GPU cluster, and i do not have any admin rights to install CUDA driver.
ThanksVasudev
Dev, please check the CUDA toolkit installation log files. Check the driver was installed correctly without errors.Also, please check that the node where you are running nvidia-smi has a GPU and is correctly installed.If error persists, I would remove all the previous CUDA drivers installed, and start a clean installation.
i have installed it and still the error persists
[vasudev@lrz-login2 BLP]$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2018 NVIDIA Corporation
Built on Sat_Aug_25_21:08:01_CDT_2018
Cuda compilation tools, release 10.0, V10.0.130
[vasudev@lrz-login2 BLP]$ nvidia-smi
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
You will need to reinstall the driver.I would install CUDA toolkit, the driver will be installed as part of it:
Dear moises,
I have checked nvidia-SMI and nvcc --version
[vasudev@lrz-login3 BLP]$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2015 NVIDIA Corporation
Built on Tue_Aug_11_14:27:32_CDT_2015
Cuda compilation tools, release 7.5, V7.5.17
[vasudev@lrz-login3 BLP]$ nvidia-smi
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
ThanksVasudev
I would use nvidia-smi to check what GPU(s) the node has, and if they are free.
Dear Moises,
Yes, same error persists, although i am running the bedpostx_gpu on GPU cluster.I dont know the reason for this problem
ThanksVasudev
I can see exactly the same error: no GPU
Dear moises,
I have reworked on analysis, but still , i am unsure if the file size that i am getting is accurate
Kindly check the results of this subject.
Thank youVasudev
Hi Vasudev,in your log files you can find this message:cuda error at CUDA/init_gpu.cu:17. no CUDA-capable device is detected
So no GPU was found (or free) on that node
Moises
Dear all,
After i run bedpostx_gpu analysis, i am confused if the size of the files merged_th1samples,merged_th2samples etc should be larger than what i am getting, could you please check the sample output ( https://www.dropbox.com/s/uby0erm3ntk14s4/KON11.bedpostX.zip?dl=0 ) and let me know, if you think bedpostx_gpu was accurately run .
ThanksVasudev
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1