Skip to content

Exercise session 1: Running applications

Exercises are in Exercises/HPE/day1/ProgrammingModels.`

See Exercises/HPE/day1/ProgrammingModels/ProgrammingModelExamples_SLURM.pdf Run on the system and get familiar with Slurm commands.

Materials

Archived materials on LUMI:

  • Exercise assignments in /appl/local/training/4day-20240423/files/LUMI-4day-20240423-Exercises_HPE.pdf

  • Exercises as bizp2-compressed tar file in /appl/local/training/4day-20240423/files/LUMI-4day-20240423-Exercises_HPE.tar.bz2

  • Exercises as uncompressed tar file in /appl/local/training/4day-20240423/files/LUMI-4day-20240423-Exercises_HPE.tar

Q&A

  1. Instead of slurm I would be interested in using myqueue to submit my scripts. Do you know if this is possible? https://myqueue.readthedocs.io/en/latest/

    • We do not recommend it but you can try it yourself. It may be tricky to configure properly and we do not offer support fot the tool. Remember you will be charged for automatically (re)submitted jobs which can run out of your control if the myqueue instance is not properly configured.

  2. I'm trying to compile the pi_hip.cpp CC -xhip -o test.hip pi_hip.cpp, BUT I am getting 26 warnings. e.g. 

    In file included from /opt/rocm/hip/include/hip/amd_detail/texture_fetch_functions.h:26:
In file included from /opt/rocm/hip/include/hip/amd_detail/../../../../include/hip/amd_detail/texture_fetch_functions.h:28:
/opt/rocm/hip/include/hip/hip_texture_types.h:25:9: warning: This file is deprecated. Use file from include path /opt/rocm-ver/include/ and prefix with hip [-W#pragma-messages]
#pragma message("This file is deprecated. Use file from include path /opt/rocm-ver/include/ and prefix with hip”)

    How to get rid of those warnings ?

    • This is the Cray compiler wrapper including the old (deprecated) path. This is harmless and can be ignored. If it really bother you, you can use the -Wno-#pragma-messages compiler flag to silence the warning.

  3. How to launch one simple command with srun, which partition and command to use?

    • You need to specify partition with -p option, standard-g for GPU nodes and standard for CPU only

    • Project account is also mandatory with -A option, project_465001098 is for the training purpose

    • You can also use reservation to use extra reserved resources, --reservation=LUMItraining_G or LUMItraining_G; this is only valid for this training

    • To get actual GPU devices allocated, use --gpus= from 1 to 8 for per node allocation

    • Use --exclusive to get the node(s) allocated for your job exclusively, if you submit to small(-g) or dev-g partitions

  4. What if I get this error when trying to utilize GPU resources?

    salloc -p dev-g -A project_465001098 -n 1 --gpus=8
salloc: Pending job allocation 6925988
salloc: job 6925988 queued and waiting for resources
salloc: job 6925988 has been allocated resources
salloc: Granted job allocation 6925988
piangate@uan01:~/prova1> rocm-smi 
cat: /sys/module/amdgpu/initstate: No such file or directory
ERROR:root:Driver not initialized (amdgpu not found in modules)

    • salloc gives you the reserved node, then you need srun to run on it (you see in the prompt that you are still on uan after the salloc). You can even open a bash on the compute node via srun -n1 --pty bash (best is to specify allocation time on salloc)

    • Try srun rocm-smi