Linux RHEL CentOS Installation - jackyko1991/gadgetron GitHub Wiki

RHEL 6 Derivatives

Installation instruction for Red Hat Enterprise Linux and its derivatives (CentOS, Scientific Linux, etc). These notes have been tested on CentOS 7 with the "Software Development Workstation" package option.

Enabling some none-default repositories

There are a few dependencies not in the default repositories, so we add the EPEL, ACE, and Boost repositories.

The steps below must be executed as a super user.

  • Add EPEL to the list of know repositories:
    • wget http://dl.fedoraproject.org/pub/epel/6/x86_64/epel-release-6-8.noarch.rpm
    • yum localinstall epel-release-6-8.noarch.rpm
  • Install ACE
    • mkdir ~/software && cd ~/software
    • wget ftp://ftp.pbone.net/mirror/ftp5.gwdg.de/pub/opensuse/repositories/devel:/libraries:/ACE:/micro/CentOS_7/x86_64/ace-6.3.3-55.1.x86_64.rpm
    • yum install ./ace-6.3.3-55.1.x86_64.rpm
    • wget ftp://ftp.pbone.net/mirror/ftp5.gwdg.de/pub/opensuse/repositories/devel:/libraries:/ACE:/micro/CentOS_7/x86_64/mpc-6.3.3-55.1.x86_64.rpm
    • yum install ./mpc-6.3.3-55.1.x86_64.rpm
    • wget ftp://ftp.pbone.net/mirror/ftp5.gwdg.de/pub/opensuse/repositories/devel:/libraries:/ACE:/micro/CentOS_7/x86_64/ace-devel-6.3.3-55.1.x86_64.rpm
    • yum install /ace-devel-6.3.3-55.1.x86_64.rpm
  • Add SLC @ CERN repository:
    • wget http://linuxsoft.cern.ch/cern/scl/slc6-scl.repo -O /etc/yum.repos.d/slc6-scl.repo

Gadgetron dependencies

The steps below must be executed as a super user.

  • yum install centos-release-scl-rh

  • yum --enablerepo=centos-sclo-rh-testing install devtoolset-7-all

  • yum group install "Development Tools"

  • yum install qt-devel fftw-devel freeglut-devel hdf5-devel glew-devel lapack-devel xerces-c-devel xsd

  • yum install docbook-utils-pdf docbook5-schemas docbook5-style-xsl

  • yum install cmake cmake3 cmake-gui cmake3-gui

  • yum install boost-devel doxygen git libxml2-devel libxslt-devel openblas-devel armadillo-devel gtest

  • yum install scl-utils

    Optional: Libraries below can be updated from third party repos:

    • fftw-devel 3.2.2: atrpms
    • libglew-devel 1.7.0: linuxtech-release

Boost

It is better to download latest boost and compile and install.

cd ~/software
wget https://dl.bintray.com/boostorg/release/1.66.0/source/boost_1_66_0.tar.gz
tar xvf ./boost_1_66_0.tar.gz
cd boost_1_66_0
./bootstrap.sh
sudo ./b2 --layout=versioned --build-type=complete address-model=64 install -j 20

Intel MKL (Optional)

cd ~/software
wget http://registrationcenter-download.intel.com/akdlm/irc_nas/tec/12414/l_mkl_2018.1.163.tgz
tar -xvf ./l_mkl_2018.1.163.tgz
cd l_mkl_2018.1.163
sudo ./install.sh`

Matlab (Optional)

  • Try Matlab at https://www.mathworks.com/programs/trials/trial_request.html
  • Require JDK to compile Matlab module
    • wget --no-check-certificate --no-cookies --header "Cookie: oraclelicense=accept-securebackup-cookie" http://download.oracle.com/otn-pub/java/jdk/8u45-b14/jdk-8u45-linux-x64.rpm
    • yum localinstall jdk-8u45-linux-x64.rpm

CUDA and GPU Driver (Optional)

The steps below must be executed as a super user.

  • Install Nvidia CUDA (7.0) Toolkit and GPU driver
    • wget http://developer.download.nvidia.com/compute/cuda/repos/rhel6/x86_64/cuda-repo-rhel6-7.0-28.x86_64.rpm
    • yum localinstall cuda-repo-rhel6-7.0-28.x86_64.rpm
    • yum install cuda

DICOM Support (Optional)

The steps below must be executed as a super user.

  • Create /etc/yum.repos.d/PUIAS_6_computational.repo and add the following lines:

    [PUIAS_6_computational] name=PUIAS computational Base $releasever - $basearch mirrorlist=http://puias.math.ias.edu/data/puias/computational/$releasever/$basearch/mirrorlist gpgcheck=1 gpgkey=file:///etc/pki/rpm-gpg/RPM-GPG-KEY-puias

  • Download and install the gpg key.

    • wget http://springdale.math.ias.edu/data/puias/6/x86_64/os/RPM-GPG-KEY-puias
    • rpm --import /etc/pki/rpm-gpg/RPM-GPG-KEY-puias
    • yum install dcmtk360-devel

  • Or, you can download and compile dcmtk

cd ~/software
wget ftp://dicom.offis.de/pub/dicom/offis/software/dcmtk/dcmtk363/dcmtk-3.6.3.tar.gz
tar xvf ./dcmtk-3.6.3.tar.gz
cd dcmtk-3.6.3
mkdir build
cd build
cmake -DBUILD_SHARED_LIBS=ON ..
make -j20
sudo make install

Python Options

sudo yum install python python-devel python-pip
sudo pip install --upgrade pip
sudo pip install setuptools
sudo pip install Cython
sudo pip install numpy 
sudo pip install pyxb psutil lxml pydicom
sudo pip install 'h5py==2.5.0'

ISMRMRD and Gadgetron

The steps may be executed as a regular user.

  • Environment variables for finding Matlab and DCMTK
    • export MATLAB_ROOT=/usr/local/MATLAB/R2013a
    • export DCMTK_HOME=/usr/local/dcmtk/3.6.0
  • Create install directory
    • mkdir ~/local
  • Switch GCC to 4.9.1
    • scl enable devtoolset-7 bash
  • ISMRMRD
    • git clone https://github.com/ismrmrd/ismrmrd.git
    • cd ismrmrd
    • mkdir build; cd build
    • cmake -D CMAKE_INSTALL_PREFIX=~/local -D CMAKE_PREFIX_PATH=~/local/ ../
    • make;make install
  • Gadgetron
    • git clone https://github.com/gadgetron/gadgetron.git
    • cd gadgetron
    • mkdir build; cd build
    • cmake -D CMAKE_INSTALL_PREFIX=~/local -D CMAKE_PREFIX_PATH=~/local/ ../
    • make; make install

Environment Variables

The final step is to add/modify a few environment variables in your ~/.bashrc file:

export GADGETRON_HOME=/usr/local/gadgetron
export ISMRMRD_HOME=/usr/local
export PATH=$PATH:$GADGETRON_HOME/bin:$ISMRMRD_HOME/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$GADGETRON_HOME/lib:$ISMRMRD_HOME/lib

Rename the example configuration file $GADGETRON_HOME/config/gadgetron.xml.example to $GADGETRON_HOME/config/gadgetron.xml.

Enable firewall

sudo firewall-cmd --get-active-zones
sudo firewall-cmd --permanent --zone=public --add-port=9002/tcp
sudo firewall-cmd --permanent --zone=public --add-port=9003/tcp
sudo firewall-cmd --permanent --zone=public --add-port=9008/tcp
sudo firewall-cmd --reload