Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
gpu [2018/03/21 15:29] ufal [Servers with GPU units] |
gpu [2019/04/17 14:11] ufal [Servers with GPU units] |
||
---|---|---|---|
Line 4: | Line 4: | ||
===== Servers with GPU units ===== | ===== Servers with GPU units ===== | ||
- | GPU cluster '' | + | GPU cluster '' |
- | | machine | GPU type | GPU driver version | [[https:// | + | | machine | GPU type | GPU driver version | [[https:// |
- | | dll1 | GeForce GTX 1080 | 384.69 | 6.1 | 8 | 8114 | 250 | | + | | dll1 | GeForce GTX 1080 | 396.24 | 6.1 | 8 | |
- | | dll2 | GeForce GTX 1080 | 387.34 | 6.1 | 7 | 8114 | 250 | | + | | dll2 | GeForce GTX 1080 | 396.24 | 6.1 | |
- | | dll3 | GeForce GTX 1080 Ti | 375.66 | 6.1 | 9 | 11172 | 250 | | + | | dll3 | GeForce GTX 1080 Ti | 396.24 | 6.1 | |
- | | dll4 | GeForce GTX 1080 Ti | 375.66 | 6.1 | 10 | 11172 | 250 | | + | | dll4 | GeForce GTX 1080 Ti | 396.24 | 6.1 | 10 | |
- | | dll5 | GeForce GTX 1080 Ti | 384.69 | 6.1 | 10 | 11172 | 250 | | + | | dll5 | GeForce GTX 1080 Ti | 396.24 | 6.1 | 10 | |
- | | dll6 | GeForce GTX 1080 Ti | 384.69 | 6.1 | 9 | 11172 | 122 | | + | | dll6 | GeForce GTX 1080 Ti | 396.24 | 6.1 | 9 | |
- | | titan | GeForce GTX 1080 | 381.22 | 6.1 | 1 | 8114 | 31 | | + | | dll7 | GeForce GTX 1080 Ti | 396.24 | 6.1 | 1 | |
- | | twister1 | + | | kronos |
- | | 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 | | + | |
+ | GPU cluster '' | ||
+ | |||
+ | | machine | GPU type | GPU driver version | [[https:// | ||
+ | | tdll1 | Quadro P5000 | 410.48 | 6.1 | 8 | 17 | 245 | | ||
+ | | tdll2 | Quadro P5000 | 410.48 | 6.1 | 8 | 17 | 249 | | ||
+ | | tdll3 | Quadro P5000 | 410.48 | 6.1 | 8 | 17 | 249 | | ||
+ | | tdll4 | Quadro P5000 | 410.48 | 6.1 | 8 | 17 | 245 | | ||
+ | | tdll5 | Quadro P5000 | 410.48 | 6.1 | 8 | 17 | 249 | | ||
Desktop machines: | Desktop machines: | ||
Line 26: | Line 30: | ||
| athena | | athena | ||
- | Not used at the moment: GeForce GTX 570 (from twister2) | + | Multiple versions of CUDA library are accessible on each machine together with cudnn. Theano and TensorFlow |
- | All machines have CUDA8.0 and should support both Theano and TensorFlow. | + | |
- | [[https:// | + | [[http:// |
===== Rules ===== | ===== Rules ===== | ||
* First, read [[internal: | * First, read [[internal: | ||
- | * 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 '' | + | * 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 '' |
- | * Always specify the number of GPU cards (e.g. '' | + | * **Note that you need to use '' |
- | * If you need more than one GPU card (on a single machine), always require as many CPU cores ('' | + | * Always specify the number of GPU cards (e.g. '' |
- | * For interactive jobs, you can use '' | + | * If you need more than one GPU card (on a single machine), always require as many CPU cores ('' |
- | * Note that the dll machines have typically 10 cards, but " | + | * For interactive jobs, you can use '' |
+ | * Note that the dll machines have typically 10 cards, but " | ||
===== How to use cluster ===== | ===== How to use cluster ===== | ||
Line 44: | Line 48: | ||
==== Set-up CUDA and CUDNN ==== | ==== Set-up CUDA and CUDNN ==== | ||
- | You should add the following commands into your ~/.bashrc | + | Multiple versions of '' |
- | | + | You need to set library path from your '' |
- | CUDA_version=8.0 | + | |
- | CUDA_DIR_OPT=/ | + | |
+ | CUDA_version=9.0 | ||
+ | CUDA_DIR_OPT=/ | ||
if [ -d " | if [ -d " | ||
CUDA_DIR=$CUDA_DIR_OPT | CUDA_DIR=$CUDA_DIR_OPT | ||
Line 58: | Line 64: | ||
fi | fi | ||
- | When not using Theano, just Tensorflow this can be simplified to '' | + | * When not using Theano, just Tensorflow this can be simplified to '' |
- | + | * Note that the '' | |
- | TensorFlow 1.5 precompiled binaries need CUDA 9.0, for this you need to | + | |
- | + | ||
- | export LD_LIBRARY_PATH=/opt/cuda-9.0/lib64/:/ | + | |
- | You also need to use '' | ||
==== TensorFlow Environment ==== | ==== TensorFlow Environment ==== | ||
- | Majority | + | Many people at UFAL use TensorFlow. To start using it it is recommended |
pip install tensorflow | pip install tensorflow | ||
Line 79: | Line 81: | ||
And then you can activate your environment: | And then you can activate your environment: | ||
- | source activate | + | source activate |
- | source activate | + | source activate |
- | This environment have TensorFlow 1.0 and all necessary requirements for NeuralMonkey. | + | This environment have TensorFlow 1.8.0 and all necessary requirements for NeuralMonkey. |
==== Pytorch Environment ==== | ==== Pytorch Environment ==== | ||
Line 96: | Line 98: | ||
As an alternative to '' | As an alternative to '' | ||
- | qsubmit --gpumem=2G --queue=" | + | qsubmit --gpumem=2G --queue=" |
| | ||
- | 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. | + | 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. Please, do not use priority between -99 to 0 for jobs taking longer than a few hours, unless it is absolutely necessary for your work. |
==== Basic commands ==== | ==== Basic commands ==== | ||
Line 137: | Line 139: | ||
| titan | GeForce GTX 1080 Ti | | | titan | GeForce GTX 1080 Ti | | ||
| dll1 | (2 GPU) GeForce GTX 1080; cc6.1 | | | dll1 | (2 GPU) GeForce GTX 1080; cc6.1 | | ||
+ | | twister2 | ||
| dll2 | GeForce GTX 1080; cc6.1 | | | dll2 | GeForce GTX 1080; cc6.1 | | ||
| titan-gpu | | titan-gpu | ||
Line 154: | Line 157: | ||
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; Cuda capability |
- | | Tesla K40c; cc3.5 | | + | | 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 |
| 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 ===== | ||
Line 171: | Line 174: | ||
==== Individual acquisitions: | ==== Individual acquisitions: | ||
- | There is an easy way to get one high-end GPU: [[https:// | + | There is an easy way to get one high-end GPU: [[https:// |
Take care, however, to coordinate the grant applications a little, so that not too many arrive from UFAL within a short time: these grants are explicitly //not// intended to build GPU clusters, they are " | Take care, however, to coordinate the grant applications a little, so that not too many arrive from UFAL within a short time: these grants are explicitly //not// intended to build GPU clusters, they are " |