Arkansas High Performace Computing Center [hpcwiki]

User Tools

Site Tools

Trace: • ssh_certificates • federated_identity_login • gromacs2023
singularity-apptainer

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
singularity-apptainer [2023/06/14 20:49]
root 		singularity-apptainer [2023/06/14 20:54] (current)
root
Line 134: Line 134:
 $ docker pull nvcr.io/nvidia/hpc-benchmarks:23.5 $ docker pull nvcr.io/nvidia/hpc-benchmarks:23.5
 $ docker save nvcr.io/nvidia/hpc-benchmarks:23.5 /mystoragelocation/hpc-benchmarks.tar $ docker save nvcr.io/nvidia/hpc-benchmarks:23.5 /mystoragelocation/hpc-benchmarks.tar
-$ scp /mystoragelocation/hpc-benchmarks.tar myhpcuserid@hpc-portal2.hpc.uark.edu:/storage/myhpcuserid/+$ scp /mystoragelocation/hpc-benchmarks.tar rfeynman@hpc-portal2.hpc.uark.edu:/storage/rfeynman/
 </code> </code>
  
-As myhpcuserid on the cluster, use srun to get a cloud node, this takes a few minutes:+As "rfeynman" on the cluster, use srun to get a cloud node, this takes a few minutes:
 <code> <code>
-$ cd /storage/myhpcuserid+$ cd /storage/rfeynman
 $ singularity build hpc-benchmarks.sif docker-archive://hpc-benchmarks.tar $ singularity build hpc-benchmarks.sif docker-archive://hpc-benchmarks.tar
 </code> </code>
  
-Then for nvidia get a gpu node with srun (and use --nv), or continue with cloud node (without --nv) for serial computing:+Then for nvidia, get a gpu node with srun (and use shell --nv), or continue with cloud node (without --nv) for serial cpu computing:
  
 <code> <code>
-$ singularity shell --nv --bind /scrfs/storage/myuserid:/mnt hpc-benchmarks.sif+$ singularity shell --nv --bind /scrfs/storage/rfeynman:/mnt hpc-benchmarks.sif
 </code>  </code> 
  
-For comparison, we'll run HPL with the cpu on the same node in bare metal. At this memory size it's about 2.4 TF for this dual 7543 AMD.+For comparison, we'll run HPL with the cpu on the same node in bare metal. At this memory size it runs about 2.4 TF for this dual 7543 AMD.
  
 <code> <code>
Line 193: Line 193:
 -------------------------------------------------------------------------------- --------------------------------------------------------------------------------
 WC32L2C4       62976   216     4     4              69.49             2.3963e+03 WC32L2C4       62976   216     4     4              69.49             2.3963e+03
-HPL_pdgesv() start time Wed Jun 14 15:42:54 2023 
  
-HPL_pdgesv() end time   Wed Jun 14 15:44:03 2023 
  
 ================================================================================ ================================================================================
Line 209: Line 207:
 </code> </code>
  
-We'll copy this to apptainer and change the MPI grid from 4x4 to 1x1 in HPL.dat.  In this measure, a single NVidia A100 GPU is a little over 5 times as fast as two AMD 7543 CPUs.  The 40 GB memory of this GPU is close to full.+We'll copy this HPL.dat to apptainer and change the MPI grid from 4x4 to 1x1.  In this measure, a single NVidia A100 GPU is a little over 5 times as fast as two AMD 7543 CPUs.  The 40 GB memory of this GPU is close to full, though the CPU memory could hold more.
  
 <code> <code>
singularity-apptainer.1686775787.txt.gz · Last modified: 2023/06/14 20:49 by root

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: GNU Free Documentation License 1.3
GNU Free Documentation License 1.3 Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki