[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki


[ Back to the navigation ]

This is an old revision of the document!


Table of Contents

GPU at ÚFAL

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.

Servers with GPU units

GPU cluster gpu.q at Malá Strana:

machine GPU type cc GPUs GPU RAM Comment
iridium Quadro K2000 cc3.0 1 2 GB
titan-gpu GeForce GTX Titan Z cc3.5 2 6 GB
twister1; twister2; kronos Tesla K40c cc3.5 1 12 GB
dll1; dll2 GeForce GTX 1080 cc6.1 8 8 GB
titan GeForce GTX 1080 cc6.1 1 8 GB
dll3; dll4; dll5 GeForce GTX 1080 Ti cc6.1 10 11 GB dll3 has only 9 GPUs since 2017/07
dll6 GeForce GTX 1080 Ti cc6.1 3 11 GB

Desktop machines:

machine GPU type cc GPUs GPU RAM Comment
victoria; arc GeForce GT 630 cc3.0 1 2 GB desktop machine
athena GeForce GTX 1080 cc6.1 1 8 GB Tom's desktop machine

Not used at the moment: GeForce GTX 570 (from twister2)
All machines have CUDA8.0 and should support both Theano and TensorFlow.

Disk space

All the GPU machines are at Malá Strana (not at Troja), so you should not use /lnet/tspec/work/, but you should use:
- /lnet/spec/work/ (alias /net/work/) - Lustre disk space at Malá Strana
- /net/cluster/TMP - NFS hard disk for temporary files, so slower than Lustre for most tasks
- /net/cluster/SSD - also NFS, but faster then TMP because of SSD
- /COMP.TMP - local (for each machine) space for temporary files (use it instead of /tmp; over-filling /COMP.TMP should not halt the system).

Individual acquisitions: NVIDIA Academic Hardware Grants

There is an easy way to get one high-end GPU: ask NVIDIA for an Academic Hardware Grant. All it takes is writing a short grant application (at most ~2 hrs of work from scratch; if you have a GAUK, ~15 minutes of copy-pasting). Due to the GPU housing issues (mainly rack space and cooling), Milan F. said we should request the Tesla-line cards (2017 check with Milan about this issue). If you want to have a look at an application, feel free to ask at hajicj@ufal.mff.cuni.cz :)

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 “seeding” grants meant for researchers to try out GPUs (and fall in love with them, and buy a cluster later). If you are planning to submit the hardware grant, have submitted one, or have already been awarded one, please add yourself here.

Known NVIDIA Academic Hardware Grants:

How to use cluster

In this section will be explained how to use cluster properly.

Set-up CUDA and CUDNN

You can add following command into your ~/.bashrc

CUDNN_version=6.0
CUDA_version=8.0
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

TensorFlow Environment

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=/a/merkur3/kocmanek/ANACONDA/bin:$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.

Using cluster

Rule number one, always use the GPU queue (never log in machine by ssh). Always use qsub or qsubmit with proper arguments.

For testing and using the cluster interactively you can use qrsh (this should not be used for long running experiments since the console is not closed on the end of the experiment). Following command will assign you a GPU and creates interactive console.

qrsh -q gpu.q -l gpu=1,gpu_ram=2G -pty yes bash

For running experiments you must use qsub command:

qsub -q gpu.q -l gpu=1,gpu_cc_min3.5=1,gpu_ram=2G WHAT_SHOULD_BE_RUN

Cleaner way to use cluster is with /home/bojar/tools/shell/qsubmit

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.

Basic commands

lspci
  # is any such hardware there?
nvidia-smi
  # more details, incl. running processes on the GPU
  # nvidia-* are typically located in /usr/bin
watch nvidia-smi
  # For monitoring GPU activity in a separate terminal (thanks to Jindrich Libovicky for this!)
nvcc --version
  # this should tell CUDA version
  # nvcc is typically installed in /usr/local/cuda/bin/
theano-test
  # dela to vubec neco uzitecneho? :-)
  # theano-* are typically located in /usr/local/bin/
/usr/local/cuda/samples/1_Utilities/deviceQuery/deviceQuery
  # shows CUDA capability etc.
  

Select GPU device

Use variable CUDA_VISIBLE_DEVICES to constrain tensorflow to compute only on the selected one. For the use of first GPU use (GPU queue do this for you):

export CUDA_VISIBLE_DEVICES=0

To list available devices, use:

/opt/cuda/samples/1_Utilities/deviceQuery/deviceQuery | grep ^Device

Performance tests

* 980 vs 1080 vs Titan X (not the Titan Z we have)

In the following table is the experiment conducted by Tom Kocmi. You can replicate experiment: /a/merkur3/kocmanek/Projects/GPUBenchmark (you will need to prepare environment of TensorFlow11 or use my ANACONDA). The benchmark uses 2GB model of seq2seq machine translation in Neural Monkey (De > EN). If not specified, the benchmark had an access only to one GPU.

machine Setup; CPU/GPU; Capability [cc] Walltime Note
athena GeForce GTX 1080; cc6.1 9:55:58 Tom's desktop
dll2 GeForce GTX 1080; cc6.1 10:19:40
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
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
kronos-dev Tesla K40c; cc3.5 22:41:01
twister2 Tesla K40c; cc3.5 22:43:10
twister1 Tesla K40c; cc3.5 24:19:45
helena1 16x cores CPU 46:33:14
belzebub 16x cores CPU 52:36:56
iridium Quadro K2000; cc3.0 59:47:58
helena7 8x cores CPU 60:39:17
arc GeForce GT 630; cc3.0 103:42:30 (approximated after 66 hours)
lucifer4 8x cores CPU 134:41:22

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 GPU RAM Walltime Batch size Machine
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

GPU specs for those GPUs we have:


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