[ 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 [2018/04/12 15:17]
kocmanek [Performance tests] 		gpu [2018/05/30 09:57]
vodrazka [Set-up CUDA and CUDNN]
Line 12: Line 12:
 | dll5 |  GeForce GTX 1080 Ti |  384.69 |  6.1 |  10 |  11 |  250 | | dll5 |  GeForce GTX 1080 Ti |  384.69 |  6.1 |  10 |  11 |  250 |
 | dll6 |  GeForce GTX 1080 Ti |  384.69 |  6.1 |  9 |  11 |  122 | | dll6 |  GeForce GTX 1080 Ti |  384.69 |  6.1 |  9 |  11 |  122 |
-| gpu |  GeForce GTX TITAN Z |  381.22 |  3.5 |  2 |  6 |  31 |+| titan-gpu |  GeForce GTX TITAN Z |  381.22 |  3.5 |  2 |  6 |  31 |
 | iridium |  Quadro K2000 |  367.48 |  3.0 |  1 |  2 |  504 | | iridium |  Quadro K2000 |  367.48 |  3.0 |  1 |  2 |  504 |
 | kronos |  GeForce GTX 1080 Ti |  384.81 |  6.1 |  1 |  11 |  125 | | kronos |  GeForce GTX 1080 Ti |  384.81 |  6.1 |  1 |  11 |  125 |
Line 64: Line 64:
 You also need to use ''qsub -q gpu.q@dll[256]'' because only those machines have drivers which support CUDA 9. You also need to use ''qsub -q gpu.q@dll[256]'' because only those machines have drivers which support CUDA 9.
  
-**Testing configuration (so far on twister2 only)**+**THE NEW CLUSTER (SGE 8.1.9)**
  
-Multiple versions of ''cuda'' and ''cudnn'' can be accessed in ''/opt''. +Multiple versions of ''cuda'' can be accessed in ''/opt/cuda''. **Compared to the old cluster there is a difference in setting the CUDA_DIR_OPT variable!!**
-System default version for both libraries is configured in ''/etc/ld.so.conf.d/cuda.conf'' as:+
  
-  /opt/cuda/lib64 +You need to set library path from your ''~/.bashrc'':
-  /opt/cuda/extras/CUPTI/lib64 +
-  /opt/cudnn/lib64 +
- +
-Actual version used depends on the link in ''/opt''. For example: +
- +
-  ls -l /opt +
-  ... +
-  lrwxrwxrwx 1 root root  8 dub  9 12:30 cuda -> cuda-9.0 +
-  lrwxrwxrwx 1 root root  9 dub  9 12:32 cudnn -> cudnn-7.1 +
-  ... +
-   +
-This means that the system is using ''cuda 9.0'' and ''cudnn 7.1''. +
- +
-If system default version does not work for you, you can set library path from your ''~/.bashrc''.+
  
 +  CUDNN_version=7.0
 +  CUDA_version=9.0
 +  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
  
 +  * When not using Theano, just Tensorflow this can be simplified to ''export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/cuda/9.0/cudnn/7.0/lib64:/opt/cuda/8.0/lib64''.
  
 +  * There is no default and you always need to set ''LD_LIBRARY_PATH'' explicitly.
  
 +  * Note that ''cudnn'' library is compiled for specific version of ''cuda''. If you need specific version of ''cudnn'', you can look in ''/opt/cuda/$CUDA_version/cudnn/'' whether it is available for given ''$CUDA_version''.
  
  
Line 164: Line 163:
 | titan      | GeForce GTX 1080 Ti                |   10:45:11 | (new result with correct CUDA version) | | titan      | GeForce GTX 1080 Ti                |   10:45:11 | (new result with correct CUDA version) |
 | dll1       | (2 GPU) GeForce GTX 1080; cc6.1    |   12:34:34 | Probably only one GPU was used | | dll1       | (2 GPU) GeForce GTX 1080; cc6.1    |   12:34:34 | Probably only one GPU was used |
 +| twister2   | Quadro P5000                       |   13:19:00 |  |
 | dll2       | GeForce GTX 1080; cc6.1            |   13:01:05 | Only one GPU was used | | dll2       | GeForce GTX 1080; cc6.1            |   13:01:05 | Only one GPU was used |
 | titan-gpu  | (2 GPU) GeForce GTX Titan Z; cc3.5 |   16:05:24 | Probably only one GPU was used | | titan-gpu  | (2 GPU) GeForce GTX Titan Z; cc3.5 |   16:05:24 | Probably only one GPU was used |
Line 181: Line 181:
 The previous benchmark only compares the speed of processing units within the GPUs and do not take into account the size of memory. Therefore I have conducted another benchmark, this time for each graphic card I have increased the batch size as much as possible so the model still could fit into the GPU (the previous benchmark model had batch size 20). This way the results should be more representative of the power for each GPU. The previous benchmark only compares the speed of processing units within the GPUs and do not take into account the size of memory. Therefore I have conducted another benchmark, this time for each graphic card I have increased the batch size as much as possible so the model still could fit into the GPU (the previous benchmark model had batch size 20). This way the results should be more representative of the power for each GPU.
  
-| GPU; Cuda capability       | GPU RAM |  Walltime | Batch size | Machine | +| GPU; Cuda capability       | GPU RAM |  Walltime | Batch size | Machine   | 
-| GeForce GTX 1080 Ti; cc6.1 |   11 GB |  00:55:56 |       2300 | dll5 | +| GeForce GTX 1080 Ti; cc6.1 |   11 GB |  00:55:56 |       2300 | dll5      | 
-| GeForce GTX 1080; cc6.1    |    8 GB |  01:10:57 |       1700 | dll1 |+| GeForce GTX 1080; cc6.1    |    8 GB |  01:10:57 |       1700 | dll1      | 
 +| Quadro P5000               |   16 GB |  01:17:00 |       3400 | twister2  |
 | GeForce GTX Titan Z; cc3.5 |    6 GB |  02:20:47 |       1100 | titan-gpu | | GeForce GTX Titan Z; cc3.5 |    6 GB |  02:20:47 |       1100 | titan-gpu |
-| Quadro K2000; cc3.0        |    2 GB |  28:15:26 |         50 | iridium |+| Quadro K2000; cc3.0        |    2 GB |  28:15:26 |         50 | iridium   |
  
 ===== Links ===== ===== Links =====

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