Introduction to MPI Tutorial.
This is a basic introduction to the MPI (Message Passing Interface) framework. It was designed for work on the SCC cluster.
Getting Started
- Login in to the tutorial room desktop. You can use your own laptop so long as you are familiar with logging into the cluster.
- Use MobaXterm to login to the cluster. If you need a login account,
see the board for instructions. Here's an example:
% ssh -X tuta1@scc1.bu.edu tuta1@scc1.bu.edu's password: Last login: Tue Sep 8 13:22:52 2015 from library-dhcp-134-77.bumc.bu.edu ******************************************************************************** This machine is governed by the University policy on ethics. http://www.bu.edu/tech/about/policies/computing-ethics/ This machine is owned and administered by Boston University. See the Research Computing web site for more information about our facilities. http://www.bu.edu/tech/about/research/ For Cluster specific documentation see: http://www.bu.edu/tech/about/research/computation/scc/ Please send questions and report problems to "help@scc.bu.edu". ********************************************************************************
- Choose a location where you'd like to do your work. If you're using a tutorial account, the home directory is fine.
- Clone the tutorial content:
% git clone /project/scv/examples/mpi/tutorials/introduction/ Initialized empty Git repository in /usr4/tutorial/tuta1/introduction/.git/
- Decide which language you will use for the examples (F77, C, C++ and Python).
- Navigate to the code examples (modify for your choice):
% cd introduction/intro/basic/ % ls C C++ F77 python % cd C++
Presentation (Single Page)
Follow this link to view the tutorial content as a single page.Presentation (Slide Show)
Follow this link to view the tutorial content as a sideshow.Building Content from source
The slides were generated usingipython notebook
. To
access all the required tools, use conda
to build an
environement defined in the conda_requirements.txt file.
The Makefile found in this directory has the instructions
required to build the html content from the ipython notebook.