Regarding using GPUs for ML workflows

[Misra]

Hello,

I am training my model using xgboost and processing data in parallel using Dask. It is running on CPU but I need to run it on GPU and enable multinode. Is there a way to do it? Also, is there a command or parameter to know the details of devices we are running (single/multi GPUs)?

Thank you

[JananiC_Intel]

Hi,

Thanks for posting in Intel forums.

In DevCloud, currently we allow every user to run two jobs, one job running on two nodes simultaneously or two jobs each running on one node. You can use the below command.

qsub -l nodes=2:gpu:ppn=2 -d . job.sh

To know the device details, you can use the basic command "lscpu " after entering into the compute node. For more information you can refer the below documentation.

https://devcloud.intel.com/oneapi/documentation/job-submission/

Try this and let us know the updates.

Regards,

Janani Chandran

[JananiC_Intel]

Hi,

Is your issue resolved? Do you have any update?

Regards,

Janani Chandran

[JananiC_Intel]

Hi,

We assume that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.

Regards,

Janani Chandran