User Tools

Site Tools


singularity

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
singularity [2018/03/06 21:23]
pwolinsk
singularity [2018/03/06 21:29]
pwolinsk
Line 121: Line 121:
 === Tensorflow Example - GPU NVIDIA container === === Tensorflow Example - GPU NVIDIA container ===
  
 +Start an interactive job on a gpu node:
 +<​code>​
 +razor-l1:​pwolinsk:​$ qsub -I -q gpu16core
 +qsub: waiting for job 3927490.sched to start
 +qsub: job 3927490.sched ready
 +
 +Currently Loaded Modulefiles:​
 +  1) os/el6
 +compute0805:​pwolinsk:​$ ​
 +</​code>​
 +
 +Clone the tensorflow example models:
 <​code>​ <​code>​
 compute0805:​pwolinsk:​$ git clone https://​github.com/​tensorflow/​models compute0805:​pwolinsk:​$ git clone https://​github.com/​tensorflow/​models
Line 129: Line 141:
 Receiving objects: 100% (12884/​12884),​ 412.34 MiB | 27.24 MiB/s, done. Receiving objects: 100% (12884/​12884),​ 412.34 MiB | 27.24 MiB/s, done.
 Resolving deltas: 100% (7276/​7276),​ done. Resolving deltas: 100% (7276/​7276),​ done.
 +</​code>​
 +
 +Load the singularity module and start a shell within the docker container.
 +<​code>​
 compute0805:​pwolinsk:​$ module load singularity compute0805:​pwolinsk:​$ module load singularity
 compute0805:​pwolinsk:​$ singularity shell  --nv /​share/​apps/​singularity/​images/​nvidia-tensorflow\:​18.01-py2-ahpcc.simg compute0805:​pwolinsk:​$ singularity shell  --nv /​share/​apps/​singularity/​images/​nvidia-tensorflow\:​18.01-py2-ahpcc.simg
Line 159: Line 175:
 Step 100 (epoch 0.12), 48.3 ms Step 100 (epoch 0.12), 48.3 ms
 .... ....
 +</​code>​
  
 +While the Tensorflow job is running inside the Singularity container, ssh into the node and verify that the GPUS are in use:
 +
 +<​code>​
 compute0805:​pwolinsk:​$ nvidia-smi ​ compute0805:​pwolinsk:​$ nvidia-smi ​
 Tue Mar  6 14:49:50 2018        Tue Mar  6 14:49:50 2018       
singularity.txt · Last modified: 2018/03/06 21:29 by pwolinsk