[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki

[ Back to the navigation ]

You are here: start » gpu

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
Next revision Both sides next revision
gpu [2017/07/28 23:14]
popel [GPU at ÚFAL] 		gpu [2017/10/11 13:07]
bojar people should not set CUDA_VISIBLE_DEVICES
Line 11: Line 11:
 | twister1; twister2; kronos | Tesla K40c          |  cc3.5 |   1|  12 GB |  | | twister1; twister2; kronos | Tesla K40c          |  cc3.5 |   1|  12 GB |  |
 | dll1; dll2                 | GeForce GTX 1080    |  cc6.1 |   8|   8 GB |  | | dll1; dll2                 | GeForce GTX 1080    |  cc6.1 |   8|   8 GB |  |
-| titan                      | GeForce GTX 1080 Ti |  cc6.1 |   1|  11 GB |  |+| titan                      | GeForce GTX 1080    |  cc6.1 |   1|   8 GB |  |
 | dll3; dll4; dll5           | GeForce GTX 1080 Ti |  cc6.1 |  10|  11 GB | dll3 has only 9 GPUs since 2017/07 | | dll3; dll4; dll5           | GeForce GTX 1080 Ti |  cc6.1 |  10|  11 GB | dll3 has only 9 GPUs since 2017/07 |
 +| dll6                       | GeForce GTX 1080 Ti |  cc6.1 |   3|  11 GB |  |
  
 Desktop machines: Desktop machines:
Line 46: Line 47:
  
 In this section will be explained how to use cluster properly.  In this section will be explained how to use cluster properly. 
 +
 +==== Set-up CUDA and CUDNN ====
 +
 +You can add following command into your ~/.bashrc
 +
 +  CUDNN_version=6.0
 +  CUDA_version=8.0
 +  CUDA_DIR_OPT=/opt/cuda-$CUDA_version
 +  if [ -d "$CUDA_DIR_OPT" ] ; then
 +    CUDA_DIR=$CUDA_DIR_OPT
 +    export CUDA_HOME=$CUDA_DIR
 +    export THEANO_FLAGS="cuda.root=$CUDA_HOME,device=gpu,floatX=float32"
 +    export PATH=$PATH:$CUDA_DIR/bin
 +    export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CUDA_DIR/cudnn/$CUDNN_version/lib64:$CUDA_DIR/lib64
 +    export CPATH=$CUDA_DIR/cudnn/$CUDNN_version/include:$CPATH
 +  fi
 +
 ==== TensorFlow Environment ==== ==== TensorFlow Environment ====
  
Line 98: Line 116:
   /usr/local/cuda/samples/1_Utilities/deviceQuery/deviceQuery   /usr/local/cuda/samples/1_Utilities/deviceQuery/deviceQuery
     # shows CUDA capability etc.     # shows CUDA capability etc.
 +  ssh dll1; ~popel/bin/gpu_allocations
 +    # who occupies which card on a given machine
          
 === Select GPU device === === Select GPU device ===
  
-Use variable CUDA_VISIBLE_DEVICES to constrain tensorflow to compute only on the selected one. For the use of first GPU use (GPU queue do this for you): +The variable CUDA_VISIBLE_DEVICES constrains tensorflow and other toolkits to compute only on the selected GPUs. **Do not set this variable yourself** (unless debugging SGE), it is set for you automatically by SGE if you ask for some GPUs (see above).
-  export CUDA_VISIBLE_DEVICES=0+
  
 To list available devices, use: To list available devices, use:

[ Back to the navigation ] [ Back to the content ]