System Requirements

These instructions assume that you intend to install OpenMDAO for normal use – that you will not be hacking on the OpenMDAO infrastructure or standard library, or developing any plugins that require compilation. If you do intend to perform some kind of OpenMDAO development, you should refer to the installation instructions in the Developer Guide because the developer environment has some additional system requirements on certain platforms.

OpenMDAO requires that the following software be installed at the system level on all platforms:

  • Python 2.6
  • NumPy (version 1.3 or higher)
  • SciPy
  • Matplotlib

Platform-Specific Requirements

We recommend that you read the entire section relating to your desired platform before installing any software.

Linux:

Our current OpenMDAO distribution for Linux is a source distribution, so to use it, you must have compilers (C and Fortran) on your system.

  • On Fedora, the names of the needed RPM packages are:
  • python-devel
  • numpy
  • scipy
  • python-matplotlib
  • gcc-gfortran
  • On Ubuntu, the .deb package names are:
  • python-dev
  • python-numpy
  • python-scipy
  • python-matplotlib
  • gfortran

Windows:

Mac OS X:

Our current OpenMDAO distribution for OS X is a source distribution, so to use it, you must have compilers (C and Fortran) on your system.

  • Xcode – It’s included on the OS X install disk, and installing it will give you access to gcc. You can

    also download a newer version from Apple, but you’ll have to fill out a (free) registration to do it.

  • gfortran – It’s sometimes hard to figure out which version of gfortran to install on your Mac. See

    this page for a pretty good overview of what’s available.

  • On Snow Leopard:

  • Python 2.6 – Install a new version from python.org because the built-in version has a distutils bug that will cause some of the OpenMDAO tests to fail. It’s not clear exactly what release the fix first appeared in, but version Python 2.6.5 definitely has the fix.
  • numpy
  • SciPy
  • gfortran – This version goes with Xcode.
  • Matplotlib
  • On Leopard:

If you have g77 installed on Leopard, you may get build errors like: “ld: library not found for -lcc_dynamic”. This indicates that g77 is being used instead of gfortran, which won’t work. At the moment, the recommended fix is to change the name of g77 to something else, for example, _g77 so that it won’t be found by numpy.distutils.

Installation

To avoid version conflicts with system level Python packages, OpenMDAO is best installed using a bootstrap script that installs OpenMDAO in a virtual Python environment. This script is called go-openmdao.py.

Acquire and run the script

  1. Download the installation script: go-openmdao.py

    If you are using Internet Explorer, right click on the above link; then select “Save Target as.” Be sure to save the go-openmdao.py script to the folder where you want to install OpenMDAO. Other browsers may automatically download the file to a specific folder (e.g., a Downloads folder), and you will have to move it to the location where you want to install OpenMDAO. We highly recommend you pick a folder without any spaces in your path name! For instance, on a Windows machine, you could put all your OpenMDAO installations in C:\\OpenMDAO.

    If you are on a Windows machine, even though you have Python installed, the Path environment variable will not be configured to point to the Python installation directory. You will have to put Python26 in the path. For help doing this, please see the following video. Please note that this video is for Python 2.5, and you want to use Python 2.6.

  2. Run the script.

    If you specify a directory name, the virtual environment will be installed there. If you don’t supply a directory name, it will default to a name based on the version, for example, openmdao-0.1.

    python go-openmdao.py

    This script will check the version of Python you are running. You must be running a version greater than or equal to 2.6 but less than 3.0. After the script completes successfully, you will have installed OpenMDAO. There are just a couple more steps to follow before you can start using it.

Activate your virtual environment

Each time before you want to use openmdao, you need to activate its virtual environment. Activation adds your virtual environment’s bin directory to the front of your system path in the current shell so that when you type python, you’ll get a Python interpreter that’s customized for your virtual environment, giving you access to everything in OpenMDAO.

Navigate into the folder that was just created by your install script.

If you are on Linux or Mac OS X, you must be running Bash. If you are in Bash, omit this step; otherwise, type:

bash

Next, type the following, making sure to include the “.” in the command:

. bin/activate

If you are on Windows, type:

Scripts\activate

Run tests to verify valid install

OpenMDAO has a large test suite which allows you to check and make sure all of the functionality of OpenMDAO will work on your system. You can run these tests yourself to double check your installation. It takes only a couple of minutes to run them all.

To run all tests, type the following:

openmdao_test

If you don’t get any errors or failed tests, everything worked. If you did have some problems, the forum is your next stop!

Now you are ready to start using OpenMDAO. When you are done using it, you can deactivate the environment by typing:

deactivate

Note

If you need an earlier version of OpenMDAO, its bootstrap script can be downloaded from: http://openmdao.org/downloads/<openmdao_version>/go-openmdao.py where <openmdao_version> is the version of OpenMDAO you want to install. You can browse through the older versions here.

OpenMDAO Home

Table Of Contents

Previous topic

Introduction to the OpenMDAO Framework

Next topic

Getting Started: A Simple Tutorial Problem

This Page