| Both sides previous revision
Previous revision
Next revision
|
Previous revision
Next revision
Both sides next revision
|
gpu [2018/01/23 23:16]
popel [Servers with GPU units] |
gpu [2018/03/21 15:29]
ufal [Servers with GPU units] |
| GPU cluster ''gpu.q'' at Malá Strana: | GPU cluster ''gpu.q'' at Malá Strana: |
| |
| | machine | GPU type | [[https://en.wikipedia.org/wiki/CUDA#GPUs_supported|cc]] | GPUs | GPU RAM | Comment | | | machine | GPU type | GPU driver version | [[https://en.wikipedia.org/wiki/CUDA#GPUs_supported|cc]] | GPU cnt | GPU RAM (MB) | machine RAM (GB)| |
| | iridium | Quadro K2000 | cc3.0 | 1| 2 GB | driver(iridium)=367.48 | | | dll1 | GeForce GTX 1080 | 384.69 | 6.1 | 8 | 8114 | 250 | |
| | titan-gpu | GeForce GTX Titan Z | cc3.5 | 2| 6 GB | driver(titan-gpu)=381.22 | | | dll2 | GeForce GTX 1080 | 387.34 | 6.1 | 7 | 8114 | 250 | |
| | twister1; twister2; kronos | Tesla K40c | cc3.5 | 1| 12 GB | driver(twister*)=367.48, driver(kronos)=384.81 | | | dll3 | GeForce GTX 1080 Ti | 375.66 | 6.1 | 9 | 11172 | 250 | |
| | dll1; dll2 | GeForce GTX 1080 | cc6.1 | 8| 8 GB | driver(dll1)=375.66, driver(dll2)=387.26 | | | dll4 | GeForce GTX 1080 Ti | 375.66 | 6.1 | 10 | 11172 | 250 | |
| | titan | GeForce GTX 1080 | cc6.1 | 1| 8 GB | driver(titan)=381.22| | | dll5 | GeForce GTX 1080 Ti | 384.69 | 6.1 | 10 | 11172 | 250 | |
| | dll4; dll5 | GeForce GTX 1080 Ti | cc6.1 | 10| 11 GB | driver(dll4)=375.66, driver(dll5)=384.69 | | | dll6 | GeForce GTX 1080 Ti | 384.69 | 6.1 | 9 | 11172 | 122 | |
| | dll3; dll6 | GeForce GTX 1080 Ti | cc6.1 | 9| 11 GB | driver(dll3)=375.66, driver(dll6)=384.69 | | | titan | GeForce GTX 1080 | 381.22 | 6.1 | 1 | 8114 | 31 | |
| | | twister1 | Tesla K40c | 367.48 | 3.5 | 1 | 11439 | 47 | |
| | | twister2 | Tesla K40c | 367.48 | 3.5 | 1 | 11439 | 47 | |
| | | titan-gpu | GeForce GTX TITAN Z | 381.22 | 3.5 | 2 | 6082 | 31 | |
| | | kronos | GeForce GTX 1080 Ti | 384.81 | 6.1 | 1 | 11172 | 125 | |
| | | iridium | Quadro K2000 | 367.48 | 3.0 | 1 | 1998 | 504 | |
| |
| Desktop machines: | Desktop machines: |
| * First, read [[internal:Linux network]] and [[:Grid]]. | * First, read [[internal:Linux network]] and [[:Grid]]. |
| * 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_vmem=12G''. | * 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_vmem=12G''. |
| * 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 you GPU memory requirements (e.g. ''gpu_ram=2G''). Thus e.g. <code>qsub -q gpu.q -l gpu=1,gpu_cc_min3.5=1,gpu_ram=2G</code> | * 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.q -l gpu=1,gpu_cc_min3.5=1,gpu_ram=2G</code> |
| * If you need more than one GPU card (on a single machine), always require as many CPU cores (''-pe smp X'') as many GPU cards you need. E.g. <code>qsub -q gpu.q -l gpu=4,gpu_cc_min3.5=1,gpu_ram=7G -pe smp 4</code> **Warning**: currently, this does not work, so you can omit the ''-pe smp X'' part. Milan Fučík is working on a fix. | * If you need more than one GPU card (on a single machine), always require as many CPU cores (''-pe smp X'') as many GPU cards you need. E.g. <code>qsub -q gpu.q -l gpu=4,gpu_cc_min3.5=1,gpu_ram=7G -pe smp 4</code> **Warning**: currently, this does not work, so you can omit the ''-pe smp X'' part. Milan Fučík is working on a fix. |
| * 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'' is necessary**, otherwise the variable ''$CUDA_VISIBLE_DEVICES'' will not be set correctly. E.g. <code>qrsh -q gpu.q -l gpu=1,gpu_ram=2G -pty yes bash</code>In general: don't reserve a GPU (as described above) without actually using it for longer time. (E.g. try separating steps which need GPU and steps which do not and execute those separately on our GPU resp. CPU cluster.) Ondřej Bojar has a script /home/bojar/tools/servers/watch_gpus for watching reserved but unused GPU on most machines which will e-mail you, but don't rely on in only. | * For interactive jobs, you can use ''qrsh'', but make sure to end your job as soon as you don't need the GPU (so don't use qrsh for long training). **Warning: ''-pty yes bash'' is necessary**, otherwise the variable ''$CUDA_VISIBLE_DEVICES'' will not be set correctly. E.g. <code>qrsh -q gpu.q -l gpu=1,gpu_ram=2G -pty yes bash</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 in only. |
| * Note that the dll machines have typically 10 cards, but "just" 250 GB RAM. 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. | * Note that the dll machines have typically 10 cards, but "just" 250 GB RAM (DLL6 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 different machine. |
| |
| ===== How to use cluster ===== | ===== How to use cluster ===== |
| |
| When not using Theano, just Tensorflow this can be simplified to ''export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/cuda-8.0/cudnn/6.0/lib64:/opt/cuda-8.0/lib64''. Note that on some machines (dll*, twister*), this is the current default even without setting LD_LIBRARY_PATH, but on other machines (kronos, titan, titan-gpu, iridium) you need to set LD_LIBRARY_PATH explicitly. | When not using Theano, just Tensorflow this can be simplified to ''export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/cuda-8.0/cudnn/6.0/lib64:/opt/cuda-8.0/lib64''. Note that on some machines (dll*, twister*), this is the current default even without setting LD_LIBRARY_PATH, but on other machines (kronos, titan, titan-gpu, iridium) you need to set LD_LIBRARY_PATH explicitly. |
| | |
| | TensorFlow 1.5 precompiled binaries need CUDA 9.0, for this you need to |
| | |
| | export LD_LIBRARY_PATH=/opt/cuda-9.0/lib64/:/opt/cuda/cudnn/7.0/lib64/ |
| | |
| | You also need to use ''qsub -q gpu.q@dll[256]'' because only those machines have drivers which support CUDA 9. |
| | |
| ==== TensorFlow Environment ==== | ==== TensorFlow Environment ==== |
| |
| qsubmit --gpumem=2G --queue="gpu.q" WHAT_SHOULD_BE_RUN | qsubmit --gpumem=2G --queue="gpu.q" WHAT_SHOULD_BE_RUN |
| | |
| It is recommended to use priority -100 if you are not rushing for the results and don't need to leap over your colleagues jobs. | It is recommended to use priority lower than the default -100 if you are not rushing for the results and don't need to leap over your colleagues jobs. |
| ==== Basic commands ==== | ==== Basic commands ==== |
| |
