Installing RRO 8.0.3 [Archived Documentation] · MRAN Skip to main content

Installing RRO 8.0.3 [Archived Documentation]

Archive Notice

These are not the latest instructions.

The following documentation describes how to install and uninstall Revolution R Open 8.0.3, which has been archived.

Take advantage of the latest version by downloading and installing Microsoft R Open 3.4.0.

Visit our Previous Releases page to learn about other past versions.


Due to the fact that these instructions have been archived, certain links on this page may no longer work over time.

Introduction

Revolution R Open (RRO) is the enhanced distribution of R from Revolution Analytics. It is a complete open source platform for statistical analysis and data science.

RRO 8.0.3 is based on (and 100% compatible with) version 3.1.3 of the R language, and includes additional capabilities for performance, reproducibility and platform support.

RRO is free to download and use. Learn more about Revolution R Open here.

System Requirements

Before you begin installing Revolution R Open 8.0.3, make sure the machine on which you will install meets the minimum hardware and software requirements defined below.

Supported Platforms. Revolution R Open is built and tested on the following 64-bit platforms:

  • Ubuntu 12.04, 14.04, 15.04
  • CentOS / Red Hat Enterprise Linux 5.8, 6.5, 7.0
  • SUSE Linux Enterprise Server (SLES) 10, 11
  • OpenSUSE 13.1
  • Mac OS X Mavericks (10.9), Yosemite (10.10)
  • Windows® 7.0 (SP 1), 8.1, 10 & Windows Server® 2008 R2 (SP1) and 2012

Processor. Processor 64-bit processor with x86-compatible architecture (such as AMD64, Intel 64, x86-64, IA-32e, EM64T, or x64 chips). ARM chips, Itanium-architecture chips (also known as IA-64), and non-Intel Macs are not supported. Multiple-core chips are recommended.

Free disk space. 250 MB

RAM. 1 GB required, 2 GB recommended.

Installing on Windows

Before installing Revolution R Open 8.0.3 , please review the following system requirements.

Side-by-Side Installs

On Windows, Revolution R Open 8.0.3 can work side-by-side with any other version of CRAN R, including past version of RRO. In order to continue working with a different version of R from a terminal window, you must specify the full path to the R executable you want to use.

After the installation of RRO 8.0.3, you can launch Rgui.exe to start interacting with R.

For example, if you installed Revolution R Open 8.0.0 and then upgraded to Revolution R Open 8.0.3, here is how you could work with them side-by-side:

  • Start RRO 8.0.3 by executing C:\Program Files\RRO\R-3.1.3\bin\x64\Rgui.exe if that is where you installed RRO 8.0.3.
  • Start RRO 8.0.0 by executing C:\Program Files\RRO\R-3.1.1\bin\x64\Rgui.exe if that is where you installed RRO 8.0.0.

How to Install

Important! Any package dependencies will be installed automatically during the installation of Revolution R Open as long as your machine has access to the internet.

To install Revolution R Open 8.0.3 on Windows:

  1. Make sure R is not running. Stop every R process prior to installing. You may need to close your IDE.

  2. Go to the directory where you downloaded the installer.

  3. Launch the installer and follow the prompts.

    Note: After installing, the default R path is updated to point to R installed with Revolution R Open 8.0.3.
    Important! The RRO installer will install the checkpoint package if you have an active internet connection. If you don't have a connection now and want to use checkpoint, then use this command later to install in R:
    install.packages("checkpoint", repos = "https://mran.revolutionanalytics.com/snapshot/2015-04-29")
  4. To configure Revolution R Open for reproducible research, you can disable the workspace save and reload feature. Learn more...

  5. To enable the optional multi-threaded performance, you must:

    Warning for RRO 8.0.2 or 8.0.3 Users:

    If the MKL math library for RRO 8.0.2 or 8.0.3 was previously installed, then you must uninstall RevoMath from within the Add or Remove Programs or Uninstall a program dialog before proceeding.

    A. Download the Math Kernel Library (MKL) for this release.

    B. Go to the directory into which you downloaded RevoMath-8.0.3.msi, the MKL utility.

    C. Launch RevoMath-8.0.3.msi and follow the prompts.

    D. Learn how to set the number of threads here.

  6. When the installation is complete, launch Rgui.exe to start interacting with R. On some platforms, you can find Rgui.exe under C:\Program Files\RRO\R-3.1.3\bin\x64\Rgui.exe.

What's Next After Install

1 Download and configure an IDE, such as RStudio.
2 If you're new to the R language, try this datacamp tutorial.
3 Check out some of these other resources.
 

Note: Revolution R Open also works with DeployR (installed separately). DeployR enables your application developers to deploy the R scripts, models, and data your R programmers have created in their applications.

Installing on Mac OS X

Before installing Revolution R Open 8.0.3, please review the following system requirements.

Side-by-Side Installs

At this time, side-by-side installations of Revolution R Open 8.0.3 and other versions of R are not supported on Mac OS X.

How to Install

Important! Any package dependencies will be installed automatically during the installation of Revolution R Open as long as your machine has access to the internet.

Revolution R Open can be installed on Mac OS X using the standard Apple installer package.

To install Revolution R Open 8.0.3 on Mac OS X:

  1. Make sure R is not running. Stop every R process prior to installing. You may need to close your IDE.

  2. Go to the directory where you downloaded the installer.

  3. Launch the installer by double-clicking the filename.

  4. In the dialog box, click Open.

  5. When the installer starts, follow the prompts.

  6. Click Install to begin the installation process.

    Important! The RRO installer will install the checkpoint package if you have an active internet connection. If you don't have a connection now and want to use checkpoint, then use this command later to install in your R terminal:
    install.packages("checkpoint", repos = "https://mran.revolutionanalytics.com/snapshot/2015-04-29")
  7. To configure Revolution R Open for reproducible research, you can disable the workspace save and reload feature. Learn more...

  8. When the installation is complete, start interacting with R in the R console. Launch the R console by clicking on the Revolution R Open program icon in the Applications folder.

What's Next After Install

1 Download and configure an IDE, such as RStudio.
2 If you're new to the R language, try this datacamp tutorial.
3 Check out some of these other resources.
 

Note: Revolution R Open also works with DeployR (installed separately). DeployR enables your application developers to deploy the R scripts, models, and data your R programmers have created in their applications.

Installing on Linux

Before installing Revolution R Open 8.0.3, please review the following system requirements.

Side-by-Side Installs

On Linux machines, Revolution R Open 8.0.3 can work side-by-side with any other version of CRAN R, including past version of RRO. During the installation of Revolution R Open, a symbolic link is created under /usr/bin/ that points to the last version of R that was installed, which in this case would be RRO 8.0.3.

After installation, you can start RRO 8.0.3 from a terminal window by typing R since it is the last one installed and is therefore under /usr/bin/.

You can also continue to work from a terminal window with other versions of R that were installed prior to Revolution R Open 8.0.3. To do so, specify the full path to the R executable you want to use.

For example, if you installed Revolution R Open 8.0.0 and then upgraded to Revolution R Open 8.0.3, you could work with both as follows:

  • Start RRO 8.0.3 by executing R since it was the last one installed.
  • Start RRO 8.0.0 by executing /usr/lib64/RRO/R-3.1.1/bin/R.
Tip: You can also manage multiple side-by-side installations of any application using the alternatives command. This command allows you create and manage symbolic links to the different installations, and thus easily refer to the installation of your choice. Example.

How to Install

Important! Any package dependencies will be installed automatically during the installation of Revolution R Open provided your machine has access to the internet.

To install Revolution R Open 8.0.3 on Ubuntu:

Alert! RRO installations from a live CD are not supported on Ubuntu.
  1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

  2. Make sure R is not running. Stop every R process prior to installing. You may need to close your IDE.

  3. Make sure the system repositories are up-to-date prior to installing Revolution R Open.  

    Note: This command will not install anything, but rather ensures that the repositories contain the latest software.

    sudo apt-get update
  4. Download the installer for your platform.

  5. Change to the directory where you downloaded the installer.

  6. Unzip the contents of the Revolution R Open installer tar file.

    Ubuntu 12.04, unzip as follows:

    tar -xzf RRO-8.0.3-Ubuntu-12.04.x86_64.tar.gz

    Ubuntu 14.04, unzip as follows:

    tar -xzf RRO-8.0.3-Ubuntu-14.04.x86_64.tar.gz

    Ubuntu 15.04, unzip as follows:

    tar -xzf RRO-8.0.3-Ubuntu-15.04.x86_64.tar.gz
  7. Change to the RRO-8.0.3 directory.

    cd RRO-8.0.3
  8. Run the Revolution R Open install script:

    ./install.sh
    Important! The RRO installer will install the checkpoint package if you have an active internet connection. If you don't have a connection now and want to use checkpoint, then use this command later to install in your R terminal:
    install.packages("checkpoint", repos = "https://mran.revolutionanalytics.com/snapshot/2015-04-29")
  9. To configure Revolution R Open for reproducible research, you can disable the workspace save and reload feature. Learn more...

  10. To enable the optional multi-threaded performance, do the following:

    A. Download the Math Kernel Library (MKL) for your platform.

    B. Go to the directory where you downloaded the installer.

    C. Unzip the RevoMath-8.0.3.tar.gz file and run the install script.

    tar -xzf RevoMath-8.0.3.tar.gz
    cd RevoMath
    ./RevoMath.sh

    D. Choose option 1 to install MKL and follow the onscreen prompts.

    E. Learn how to set the number of threads here.

  11. When the installation is complete, run R in a terminal to begin interacting with R.

    Note: After installing, the default R path is updated to point to R installed with Revolution R Open 8.0.3.

To install Revolution R Open 8.0.3 on CentOS or Red Hat:

  1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

  2. Make sure R is not running. Stop every R process prior to installing. You may need to close your IDE.

  3. Make sure the system repositories are up-to-date prior to installing Revolution R Open.  

    Note: This command will not install anything, but rather ensures that the repositories contain the latest software.

    sudo yum clean all
  4. Download the installer for your platform.

  5. Change to the directory where you downloaded the installer.

  6. Unzip the contents of the Revolution R Open installer tar file:

    CentOS / Red Hat 5.8, unzip as follows:

    tar -xzf  RRO-8.0.3-el5.x86_64.tar.gz

    CentOS / Red Hat 6.5, unzip as follows:

    tar -xzf RRO-8.0.3-el6.x86_64.tar.gz

    CentOS / Red Hat 7.0, unzip as follows:

    tar -xzf   RRO-8.0.3-el7.x86_64.tar.gz
  7. Change to the RRO-8.0.3 directory.

    cd RRO-8.0.3
  8. Run the Revolution R Open install script:

    ./install.sh
  9. To enable the optional multi-threaded performance, do the following:

    A. Download the Math Kernel Library (MKL) for your platform.

    B. Go to the directory where you downloaded the installer.

    C. Unzip the RevoMath-8.0.3.tar.gz file and run the install script.

    tar -xzf RevoMath-8.0.3.tar.gz
    cd RevoMath
    ./RevoMath.sh

    D. Choose option 1 to install MKL and follow the onscreen prompts.

  10. When the installation is complete, run R in a terminal to begin interacting with R.

    Note: After installing, the default R path is updated to point to R installed with Revolution R Open 8.0.3.

To install Revolution R Open 8.0.3 on SLES or openSUSE:

  1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

  2. Make sure R is not running. Stop every R process prior to installing. You may need to close your IDE.

  3. Make sure the system repositories are up-to-date prior to installing Revolution R Open.  

    Note: This command will not install anything, but rather ensures that the repositories contain the latest software.

    sudo zypper clean --all
  4. Download the installer for your platform.

  5. Change to the directory where you downloaded the installer.

  6. Unzip the contents of the installer tar file and install Revolution R Open:

    For SLES 10, unzip as follows:

    tar -xzf RRO-8.0.3-SLES10-1.1.x86_64.tar.gz

    For SLES 11, unzip as follows:

    tar -xzf RRO-8.0.3-SLES11-1.1.x86_64.tar.gz

    For openSUSE 13.1, unzip as follows:

    tar -xzf RRO-8.0.3-openSUSE-13.1.x86_64.tar.gz
  7. Install Revolution R Open:

    cd RRO-8.0.3
    ./install.sh
  8. To enable the optional multi-threaded performance, do the following:

    A. Log in as `root` or a user with `sudo` privileges. If the latter, precede commands requiring `root` privileges with `sudo`.

    B. Download the Math Kernel Library (MKL) for your platform.

    C. Go to the directory where you downloaded the installer.

    D. Unzip the RevoMath-8.0.3.tar.gz file and run the MKL install script.

         For SLES 10, unzip as follows:

       tar -xzf RevoMath-8.0.3-SLES10.tar.gz
       cd RevoMath
       ./RevoMath.sh

         For SLES 11 or openSUSE 13.1, unzip as follows:

       tar -xzf RevoMath-8.0.3.tar.gz
       cd RevoMath
       ./RevoMath.sh

    D. Choose option 1 to install MKL and follow the onscreen prompts.

  9. When the installation is complete, run R in a terminal to begin interacting with R.

    Note: After installing, the default R path is updated to point to R installed with Revolution R Open 8.0.3.

What's Next After Install

1 Download and configure an IDE, such as RStudio.
2 If you're new to the R language, try this datacamp tutorial.
3 Check out some of these other resources.
 

Note: Revolution R Open also works with DeployR (installed separately). DeployR enables your application developers to deploy the R scripts, models, and data your R programmers have created in their applications.

Installing on Cloudera

You can install RRO for Cloudera on every node of your Cloudera Hadoop cluster provided that:

  1. The Cloudera Hadoop cluster is managed by Cloudera Manager

  2. Cloudera was installed using a Cloudera Manager parcel


To install RRO for Cloudera on CentOS or Red Hat:

  1. On each node machine, install any missing package dependencies for Revolution R Open as follows:

    1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

    2. Make sure the system repositories are up-to-date prior to installing Revolution R Open. This command will not install anything, but rather ensures that the repositories contain the latest software.

      sudo yum clean all
    3. Check if the following packages (make, gcc, and gfortran) were installed:

      #check for packages
      yum list make gcc gcc-gfortran
    4. Run only the commands if one or more packages were not found during the preceding step.

      #if missing, install make
      yum install make
      
      #if missing, install gcc
      yum install gcc
      
      #if missing, install gcc-gfortran
      yum install gcc-gfortran
  2. On one machine, download the RRO for Cloudera parcel bundle. This bundle contains two files: the .parcel file and its .sha file.

  3. Change to the directory where you downloaded the RRO for Cloudera parcel bundle, RRO-8.0.3-Cloudera-el6.x86_64.tar.gz.

  4. Untar and copy these two files into your local parcel-repo, which is typically /opt/cloudera/parcel-repo.

    tar -xzf RRO-8.0.3-Cloudera-el6.x86_64.tar.gz
    cd /opt/cloudera/parcel-repo
  5. Modify permissions and ownership of the parcel file.

    chown root:root /opt/cloudera/parcel-repo/RRO-8.0.3-2-el6.parcel*
    chmod 755 /opt/cloudera/parcel-repo/RRO-8.0.3-2-el6.parcel*
  6. In a web browser, open Cloudera Manager and do the following:

    1. In the top navigation bar, click Hosts in the upper navigation bar. The All Hosts page appears.

    2. Click Parcels. The Parcels page appears.

    3. Click Check for New Parcels. RRO 8.0.3 appears.
      Tip: You may need to click All Clusters under the Location section on the left to see the new parcels.

    4. Click the Distribute for RRO in order to distribute Revolution R Open to all of the nodes of your cluster. Once the distribution process has finished, the Activate button appears.

    5. Click Activate to prepare Revolution R Open for cluster usage.

Uninstalling RRO

To uninstall Revolution R Open 8.0.3, select the tab for the relevant operating system and follow the instructions provided.

To uninstall Revolution R Open on Ubuntu:

  1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

  2. Run the following command to uninstall:

    dpkg --purge rro-8.0.3

To uninstall MKL, used for multi-threaded performance:

  1. Go to the RevoMath directory and run the MKL utility:

    cd RevoMath
    ./RevoMath.sh
  2. Choose option 2 to uninstall MKL.

To uninstall Revolution R Open on CentOS or Red Hat:

  1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

  2. Run the following command to uninstall:

    yum erase RRO-8.0.3

To uninstall MKL, used for multi-threaded performance:

  1. Go to the RevoMath directory and run the MKL utility:

    cd RevoMath
    ./RevoMath.sh
  2. Choose option 2 to uninstall MKL.

To uninstall Revolution R Open on openSUSE:

  1. Log in as root or a user with sudo privileges. If the latter, precede commands requiring root privileges with sudo.

  2. Run the following command to uninstall:

    zypper remove RRO-8.0.3

To uninstall MKL, used for multi-threaded performance:

  1. Go to the RevoMath directory and run the MKL utility:

    cd RevoMath
    ./RevoMath.sh
  2. Choose option 2 to uninstall MKL.

To uninstall Revolution R Open on Mac OS X:

Run the following commands to uninstall:

   cd /Applications
   sudo rm -rf "Revolution R Open".app
   cd /Library/Frameworks
   sudo rm -rf R.framework
   sudo rm /usr/bin/R
   sudo rm /usr/bin/Rscript

First, uninstall MKL Math Libraries:

This will revert to the original Rblas.dll and Rlapack.dll files.

  1. Remove RevoMath using the Windows instructions for uninstalling a program specific to your version of Windows. For example, on Windows 8, choose Control Panel > Programs & Features > Uninstall.

  2. Find RevoMath in the list and uninstall.

Then, uninstall Revolution R Open:

  1. Remove Revolution R Open using the Windows instructions for uninstalling a program specific to your version of Windows. For example, on Windows 8, choose Control Panel > Programs & Features > Uninstall.

  2. Find RRO for Windows 8.0.3 in the list and uninstall.

  3. Remove the directory in which you installed Revolution R Open, by default C:/Program Files/RRO.