Differences

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

--- gpu [2020/07/27 13:12]
dusek [Servers with GPU units]
+++ gpu [2024/10/02 15:21] (current)
popel
@@ Line 2: / Line 2: @@
 This page summarizes which UFAL servers have some GPU card, and suggests basic diagnostic commands, paths to installed tools, etc., simply everything necessary at the very beginning of using GPUs for experiments.
+**TODO: IN 2022 MOVING FROM SGE TO SLURM** (see [[slurm|slurm guidelines]]) -- **commands like ''qsub'' and ''qstat'' will stop working!**
+**IN 2024: Newly, all the documentation is at a dedicated wiki https://ufal.mff.cuni.cz/lrc (you need to use ufal and [[internal:welcome-at-ufal#small-linguistic-password|small-linguistic password]] to access the wiki from outside of the UFAL network).***
 ===== Servers with GPU units =====
@@ Line 7: / Line 11: @@
 | machine | GPU type | GPU driver version | [[https://en.wikipedia.org/wiki/CUDA#GPUs_supported|cc]] | GPU cnt | GPU RAM (GB) | machine RAM (GB)|
-| dll1 | Quadro RTX 5000      |  440.33 |  7.5 |   8 |  16.0 |  366.0 |
+| dll1 |  Quadro RTX 5000 |   455.23.05 |  7.5 |  8 |  16 |  366.2 |
-| dll3 |  GeForce GTX 1080 Ti |  440.33 |  6.1 |  10 |  11.0 |  248.0 |
+| dll3 |  NVIDIA RTX A4000 |   510.73.08 |  8.6 |  10 |  16 |  248.8 |
-| dll4 |  GeForce GTX 1080 Ti |  440.33 |  6.1 |  10 |  11.0 |  248.0 |
+| dll4 |  GeForce GTX 1080 Ti |   455.23.05 |  6.1 |  10 |  11 |  248.8 |
-| dll5 |  GeForce GTX 1080 Ti |  440.33 |  6.1 |  10 |  11.0 |  248.0 |
+| dll5 |  GeForce GTX 1080 Ti |   455.23.05 |  6.1 |  10 |  11 |  248.8 |
-| dll6 |  GeForce GTX 1080 Ti |  440.33 |  6.1 |   9 |  11.0 |  248.0 |
+| dll6 |  NVIDIA RTX A4000 |   510.73.08 |  8.6 |  8 |  16 |  248.8 |
-| dll7 |  GeForce RTX 2080 Ti |  418.39 |  7.5 |   8 |  11.0 |  248.0 |
+| dll7 |  NVIDIA RTX A4000 |   510.73.08 |  8.6 |  8 |  16 |  248.8 |
-| dll8 | Quadro RTX 5000      |  440.33 |  7.5 |   8 |  16.0 |  366.0 |
+| dll8 |  Quadro RTX 5000 |   455.23.05 |  7.5 |  8 |  16 |  366.2 |
-| kronos |  Tesla K40c        |  418.39 |  3.5 |   1 |  11.0 |  122.0 |
+| dll9 |  GeForce RTX 3090 |   455.23.05 |  8.6 |  4 |  25 |  183.0 |
+| dll10 |  GeForce RTX 3090 |   455.23.05 |  8.6 |  4 |  25 |  183.0 |
 GPU cluster ''gpu-troja.q'' at Troja:
 | machine | GPU type | GPU driver version | [[https://en.wikipedia.org/wiki/CUDA#GPUs_supported|cc]] | GPU cnt | GPU RAM (GB) | machine RAM (GB)|
-| tdll1 |  Quadro P5000 |  410.48 |  6.1 |  8 |  16.0 |  245.0 |
+| tdll1 |  Quadro P5000 |   455.23.05 |  6.1 |  8 |  16 |  245.0 |
-| tdll2 |  Quadro P5000 |  410.48 |  6.1 |  8 |  16.0 |  245.0 |
+| tdll2 |  Quadro P5000 |   455.23.05 |  6.1 |  8 |  16 |  245.0 |
-| tdll3 |  Quadro P5000 |  410.48 |  6.1 |  8 |  16.0 |  245.0 |
+| tdll3 |  Quadro P5000 |   455.23.05 |  6.1 |  8 |  16 |  245.0 |
-| tdll4 |  Quadro P5000 |  410.48 |  6.1 |  8 |  16.0 |  245.0 |
+| tdll4 |  Quadro P5000 |   455.23.05 |  6.1 |  8 |  16 |  245.0 |
-| tdll5 |  Quadro P5000 |  410.48 |  6.1 |  8 |  16.0 |  245.0 |
+| tdll5 |  Quadro P5000 |   455.23.05 |  6.1 |  8 |  16 |  245.0 |
 Desktop machines:
@@ Line 39: / Line 44: @@
   * All the rules from [[:Grid]] apply, even more strictly than for CPU because there are too many GPU users and not as many GPUs available. So as a reminder: always use GPUs via ''qsub'' (or ''qrsh''), never via ''ssh''. You can ssh to any machine e.g. to run ''nvidia-smi'' or ''htop'', but not to start computing on GPU. Don't forget to specify you RAM requirements with e.g. ''-l mem_free=8G,act_mem_free=8G,h_data=12G''.
     * **Note that you need to use ''h_data'' instead of ''h_vmem'' for GPU jobs.** CUDA driver allocates a lot of "unused" virtual memory (tens of GB per card), which is counted in ''h_vmem'', but not in ''h_data''. All usual allocations (''malloc'', ''new'', Python allocations) seem to be included in ''h_data''.
-  * Always specify the number of GPU cards (e.g. ''gpu=1''), the minimal Cuda capability you need (e.g. ''gpu_cc_min3.5=1'') and your GPU memory requirements (e.g. ''gpu_ram=2G''). Thus e.g. <code>qsub -q 'gpu*' -l gpu=1,gpu_cc_min3.5=1,gpu_ram=2G</code>
+  * Always specify the number of GPU cards (e.g. ''gpu=1'') and your GPU memory requirements (e.g. ''gpu_ram=2G''). Thus e.g. <code>qsub -q 'gpu*' -l gpu=1,gpu_ram=2G</code>
-  * If you need more than one GPU card (on a single machine), always require at least as many CPU cores (''-pe smp X'') as many GPU cards you need. E.g. <code>qsub -q 'gpu*' -l gpu=4,gpu_cc_min3.5=1,gpu_ram=7G -pe smp 4</code>
+  * If you need more than one GPU card (on a single machine), always require at least as many CPU cores (''-pe smp X'') as many GPU cards you need. E.g. <code>qsub -q 'gpu*' -l gpu=4,gpu_ram=7G -pe smp 4</code>
   * For interactive jobs, you can use ''qrsh'', but make sure to end your job as soon as you don't need the GPU (so don't use qrsh for long training). **Warning: ''-pty yes bash -l'' is necessary**, otherwise the variable ''$CUDA_VISIBLE_DEVICES'' will not be set correctly. E.g. <code>qrsh -q 'gpu*' -l gpu=1,gpu_ram=2G -pty yes bash -l</code>In general: don't reserve a GPU (as described above) without actually using it for longer time. (E.g. try separating steps which need GPU and steps which do not and execute those separately on our GPU resp. CPU cluster.) Ondřej Bojar has a script /home/bojar/tools/servers/watch_gpus for watching reserved but unused GPU on most machines which will e-mail you, but don't rely on it only.
   * Note that the dll machines have typically 10 cards, but "just" 250 GB RAM (DLL7 has only 128 GB). So the expected (maximal) ''mem_free'' requirement for jobs with 1 GPU is 25GB. If your 1-GPU job takes e.g. 80 GB and you submit three such jobs on the same machine, you have effectively blocked the whole machine and seven GPUs remain unused. If you really need to submit more high-memory jobs, send each on a different machine.
@@ Line 54: / Line 59: @@
 You need to set library path from your ''~/.bashrc'':
-  CUDA_version=10.1
+  CUDA_version=11.1
-  CUDNN_version=7.6
+  CUDNN_version=8.0
   CUDA_DIR_OPT=/opt/cuda/$CUDA_version
   if [ -d "$CUDA_DIR_OPT" ] ; then
@@ Line 66: / Line 71: @@
   fi
-  * When not using Theano, just Tensorflow this can be simplified to ''export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/cuda/10.1/lib64:/opt/cuda/10.1/cudnn/7.6/lib64''.
+  * When not using Theano, just Tensorflow this can be simplified to ''export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/cuda/10.1/lib64:/opt/cuda/11.1/cudnn/8.0/lib64''.
   * Note that the ''cudnn'' library is compiled for specific version of ''cuda''. If you need specific version of ''cudnn'', you can look in ''/opt/cuda/$CUDA_version/cudnn/'' whether it is available for a given ''$CUDA_version''.
@@ Line 88: / Line 93: @@
 This environment have TensorFlow 1.8.0 and all necessary requirements for NeuralMonkey.
-==== Pytorch Environment ====
+==== PyTorch Environment ====
-If you want to use pytorch, there is a ready-made environment in
+Install PyTorch following the instructions on https://pytorch.org.
-  /home/hajicj/anaconda3/envs/pytorch/bin
+At the time of writing, the recommended setup for CUDA 11.1 (supported by all GPU nodes) is:
-It does rely on the CUDA and CuDNN setup above.
+   pip3 install torch==1.10.0+cu111 torchvision==0.11.1+cu111 torchaudio==0.10.0+cu111 -f https://download.pytorch.org/whl/cu111/torch_stable.html
 ==== Using cluster ====
@@ Line 112: / Line 116: @@
   watch nvidia-smi
     # For monitoring GPU activity in a separate terminal (thanks to Jindrich Libovicky for this!)
+    # You can also use nvidia-smi -l TIME
   nvcc --version
     # this should tell CUDA version

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

Institute of Formal and Applied Linguistics Wiki

Differences