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 [2016/06/09 08:43] bojar future plans |
gpu [2018/01/08 13:25] popel dll6 now has 9 cards, so it is the same config as dll3 |
||
---|---|---|---|
Line 4: | Line 4: | ||
===== Servers with GPU units ===== | ===== Servers with GPU units ===== | ||
+ | GPU cluster '' | ||
- | | machine | GPU; [[https:// | + | | machine |
- | | titan-gpu | + | | iridium |
- | | twister1 | + | | titan-gpu |
- | | twister2 | + | | twister1; twister2; kronos |
- | | kronos-dev | + | | dll1; dll2 | GeForce GTX 1080 |
- | | kronos-dev | Tesla K40c; cc3.5? | + | | titan |
- | | //now unused// | Quadro K2000; cc3.0 | + | | dll4; dll5 | GeForce GTX 1080 Ti | |
+ | | dll3; dll6 | GeForce GTX 1080 Ti | | ||
+ | Desktop machines: | ||
+ | | machine | ||
+ | | victoria; arc | GeForce GT 630 | cc3.0 | 1 | 2 GB | desktop machine | | ||
+ | | athena | ||
- | Summary of future plans: | + | Not used at the moment: GeForce GTX 570 (from twister2) |
- | * Current Troja servers won't get any GPUs (the only option would be [[http:// | + | All machines |
- | * The old Quadro K2000 we have is a much more low end piece, so we can't test is in Troja. | + | |
- | * There is MetaCentrum which also has GPUs, so testing can be done there. | + | |
- | * It is impossible (wasteful in terms of space and forbidden by a dean regulation) to put non-rack machines to our servers rooms. So we won't be buying | + | |
- | * Yes, there are grant applications under review which include rack machines | + | |
- | Performance tests: | + | [[https://ufaladm2.ufal.hide.ms.mff.cuni.cz/ |
- | * [[http://www.trustedreviews.com/nvidia-geforce-gtx-1080-review-performance-benchmarks-and-conclusion-page-2| 980 vs 1080 vs Titan X (not the Titan Z we have)]] | + | |
- | ===== Installed toolkits ===== | ||
- | //This should mention where each interesting toolkit lives (on a particular | + | ===== Rules ===== |
+ | * 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 '' | ||
+ | * Always specify the number of GPU cards (e.g. '' | ||
+ | * If you need more than one GPU card (on a single | ||
+ | * For interactive jobs, you can use '' | ||
+ | * Note that the dll machines have typically 10 cards, but " | ||
- | ==== TensorFlow | + | ===== How to use cluster ===== |
- | [[https:// | + | ==== Set-up CUDA and CUDNN ==== |
- | OP: I created [[https:// | + | You can add following command into your ~/.bashrc |
- | === Select GPU device === | + | CUDNN_version=6.0 |
+ | CUDA_version=8.0 | ||
+ | CUDA_DIR_OPT=/ | ||
+ | if [ -d " | ||
+ | CUDA_DIR=$CUDA_DIR_OPT | ||
+ | export CUDA_HOME=$CUDA_DIR | ||
+ | export THEANO_FLAGS=" | ||
+ | export PATH=$PATH: | ||
+ | export LD_LIBRARY_PATH=$LD_LIBRARY_PATH: | ||
+ | export CPATH=$CUDA_DIR/ | ||
+ | fi | ||
- | Use variable CUDA_VISIBLE_DEVICES to constrain tensorflow to compute only on the selected one. For the use of first GPU use: | + | ==== TensorFlow Environment ==== |
- | < | + | |
- | ===== Basic commands | + | Majority people at UFAL use TensorFlow. To start using it you need to create python virtual environment (virtualenv or use Anaconda for it). Into the environment you must place TensorFlow. The TF is either in CPU or GPU version. |
+ | |||
+ | pip install tensorflow | ||
+ | pip install tensorflow-gpu | ||
+ | |||
+ | You can use prepared environment by adding into your ~/.bashrc | ||
+ | |||
+ | export PATH=/ | ||
+ | |||
+ | And then you can activate your environment: | ||
+ | |||
+ | source activate tf1 | ||
+ | source activate tf1cpu | ||
+ | |||
+ | This environment have TensorFlow 1.0 and all necessary requirements for NeuralMonkey. | ||
+ | |||
+ | ==== Pytorch Environment ==== | ||
+ | |||
+ | If you want to use pytorch, there is a ready-made environment in | ||
+ | |||
+ | / | ||
+ | |||
+ | It does rely on the CUDA and CuDNN setup above. | ||
+ | |||
+ | ==== Using cluster ==== | ||
+ | |||
+ | As an alternative to '' | ||
+ | |||
+ | qsubmit --gpumem=2G --queue=" | ||
+ | |||
+ | 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. | ||
+ | ==== Basic commands ==== | ||
lspci | lspci | ||
Line 54: | Line 100: | ||
# dela to vubec neco uzitecneho? :-) | # dela to vubec neco uzitecneho? :-) | ||
# theano-* are typically located in / | # theano-* are typically located in / | ||
+ | / | ||
+ | # shows CUDA capability etc. | ||
+ | ssh dll1; ~popel/ | ||
+ | # who occupies which card on a given machine | ||
+ | | ||
+ | === Select GPU device === | ||
+ | |||
+ | 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). | ||
+ | |||
+ | To list available devices, use: | ||
+ | / | ||
+ | |||
+ | ===== Performance tests ===== | ||
+ | |||
+ | * [[http:// | ||
+ | |||
+ | In the following table is the experiment conducted by Tom Kocmi. You can replicate experiment: / | ||
+ | |||
+ | | machine | Setup; CPU/GPU; [[https:// | ||
+ | | athena | ||
+ | | dll2 | GeForce GTX 1080; cc6.1 | | ||
+ | | titan | GeForce GTX 1080 Ti | | ||
+ | | dll1 | (2 GPU) GeForce GTX 1080; cc6.1 | | ||
+ | | dll2 | GeForce GTX 1080; cc6.1 | | ||
+ | | titan-gpu | ||
+ | | kronos-dev | Tesla K40c; cc3.5 | | ||
+ | | twister2 | ||
+ | | twister1 | ||
+ | | helena1 | ||
+ | | belzebub | ||
+ | | iridium | ||
+ | | helena7 | ||
+ | | arc | GeForce GT 630; cc3.0 | 103:42:30 | (approximated after 66 hours) | | ||
+ | | lucifer4 | ||
+ | |||
+ | |||
+ | === Second Benchmark === | ||
+ | |||
+ | 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 | ||
+ | | Tesla K40c; cc3.5 | 12 GB | | ||
+ | | 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 Titan Z; cc3.5 | 6 GB | 02:20:47 | 1100 | titan-gpu | | ||
+ | | Quadro K2000; cc3.0 | 2 GB | 28:15:26 | 50 | iridium | | ||
===== Links ===== | ===== Links ===== | ||
Line 62: | Line 154: | ||
GPU specs for those GPUs we have: | GPU specs for those GPUs we have: | ||
* [[http:// | * [[http:// | ||
+ | |||
+ | ==== Individual acquisitions: | ||
+ | |||
+ | 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 " | ||
+ | |||
+ | Known NVIDIA Academic Hardware Grants: | ||
+ | |||
+ | * Ondřej Plátek - granted (2015) | ||
+ | * Jan Hajič jr. - granted (early 2016) | ||
+ | |||
+ | |||
+ |