Does DevCloud supports Multinode for which all GPUs ?

[shubhamjain1]

As recently I got to know that Dev Cloud doesn't support Multinode for Artic Sound GPU, so wanted to understand does it support Multinode on any other GPUs.

[AthiraM_Intel]

Hi,

Thanks for posting in Intel Communities.

Could you please let us know which DevCloud you are using? (DevCloud for oneAPI/Edge/FPGA)

Thanks

[shubhamjain1]

OneAPI

[AthiraM_Intel]

Hi,

You can access multiple gpu nodes in DevCloud using the below command:

qsub -I -l nodes=2:gpu:ppn=2 -d .

If you want to list the nodes allocated to you, please use below command:

echo $PBS_NODEFILE

After this you will get a path. You can open that file with cat command.

Please find the below screenshot for reference.

If you have any further issue, kindly let us know.

Thanks

[AthiraM_Intel]

Hi,

Could you please give us an update? Is your issue resolved?

Thanks

[shubhamjain1]

No. My issue isn’t resolved. I was trying to run Multinode on xgboost and I cannot run it. I don’t know if XGboost supports Multinode on Dev Cloud

[AthiraM_Intel]

Hi,

For multi-node training with Intel XGBoost in DevCloud follow the steps mentioned in the below article.

https://medium.com/intel-analytics-software/distributed-xgboost-with-modin-on-ray-fc17edef7720

In DevCloud, multi-node computation is available through the job queue.

Syntax for multi-node:

qsub -l nodes=<count>:ppn=2

You can combine a request for multiple nodes with a request for their specific features.

Hope this helps.

Thanks

[AthiraM_Intel]

Hi,

We have not heard back from you. Could you please give us an update?

Thanks

[AthiraM_Intel]

Hi,

We have not heard back from you. This thread will no longer be monitored by Intel. If you need further assistance, please post a new question.

Thanks.