Installation

Important: Please read the instructions in this page in order. Effective use of this software is dependent on correct installation of all dependencies. Since there are a collection of dependencies, we provide tools for automating the process. So read through the instructions carefully. For a list of dependecies, see this wiki page

If you're running your version of the project on an instance of the Jetstream cloud computer, use the image "Ubuntu20.04_Anacin-X" and skip ahead to the subsection below about installing ANACIN-X. If you're using a different Jetstream image or if you're not using Jetstream, continue reading here to ensure all dependencies get installed.

If you haven't already, you will need to install the Spack and Conda package managers, be sure to do so using the following instructions. We will use these to help automate the installation of dependencies.

Spack

Spack is a package manager with good support for scientific/HPC software. To use Spack you will need Python. We recommend you install Spack and enable Spack's shell integration.

To install Spack, follow the instructions at: Spack Install

In particular, make sure to follow the instructions under "Shell support". This step will allow software installed with Spack to be loaded and unloaded as environment modules.

Conda

Conda is a cross-language package, dependency, and environment manager. We use Conda to manage the dependencies of ANACIN-X's Python code. We recommend using the Anaconda installation of the package manager rather than the Miniconda version so that you can have Jupyter installed on your machine for data visualization.

To install Conda, follow the instructions at: Conda Install

Before continuing, make sure to activate your version of conda:

source ~/.bashrc

ANACIN-X

You will need to first download the project through git using

git clone https://github.com/TauferLab/ANACIN-X.git

Note that Spack and Conda will need to be installed and set up beforehand as described above, unless you're using the Jetstream image listed above. If there is a specific C compiler that you wish to have used on your machine, please see the 'Special Case' section below.

Now that your environment is prepared and the project is on your local machine, be sure to enter the project root for setup with the command:

cd ANACIN-X

If you're using the Jetstream cloud computer image for Anacin-X titled "Ubuntu20.04_Anacin-X", you can skip this next command. Otherwise, we strongly recommend installing the dependencies for the project with:

. setup_deps.sh

The script will begin by verifying some information. Follow the prompts at the beginning, and then the installation will run on its own. The installation of all dependencies may take some time to complete.

If the 'setup_deps.sh' script did not work for you, you may need to install the dependencies manually. If that's the case, see the dependencies wiki page and install all packages that are not 'Submodule Packages' using relevant package managers.

Assuming all dependenices are installed and loaded, you will build all of ANACIN-X's components by running

. setup.sh -c

If you do not wish to build ANACIN-X with callstack tracing functionality, remove the '-c' in the above command.

Special Case

If you have a specific external C compiler installed that you wish to use, you will need to edit the compilers.yaml file in spack:

• First open the compilers.yaml file to determine which compilers are added to the system. This can be done with the command:

spack config edit compilers

• If you don't see information for your compiler in the listed options, exit the file and make sure your compiler is installed and loaded. Then use the following command before reopening the compilers.yaml file.

spack compiler find

• Within the compilers.yaml file, delete information for all compilers other than the one you wish to use.
• Save and quit the file, then try to install the ANACIN-X dependencies.