Larger computing systems such as clusters are composed of multiple shared memory nodes, connected via network infrastructures such as Infiniband. The predominant programming model used to develop parallel applications for such systems is the Message Passing Interface (MPI) standard. MPI has the advantage of supporting basically all parallel computer architectures except accelerators and is thus often considered to be the most widely used parallel programming paradigm. Over the course of three days, the participants will learn the programming model and API of MPI, and develop basic parallel applications.
This course has been originally developed by Rolf Rabenseifner from the High-Performance Computing-Center Stuttgart (HLRS) and is organized by the Research Center High Performance Computing in cooperation with HLRS and the Vienna Scientific Cluster (VSC). (Content Level: 70% for beginners, 30% advanced)
For further information and registration see: https://www.uibk.ac.at/fz-hpc/courses/mpi.html.en
Each participant will get a paper copy of all slides.
If you want, you may also buy the MPI-3.1 Standard (hardcover, 17 Euro, cash only).
Philipp Gschwandtner, philipp.gschwandtneruibk.ac.at