Coursework for MI-PDP course at FIT CTU

mi-pdp-jes is the shortest path on chess board solver, written as BB-DFS algorithm to be parallelized later.

The algorithm is thereafter parallelized using OpenMP library, for parallelization on a single computing node, and using MPI library for parallelization across many computing nodes. The program is then run on a cluster.

The original assignment is written in czech.

Assignment

• OpenMP library
• MPI library

Compile using make. Default target compiles with clang, target dev compiles without optimizations, atstar target is modified for compiling at star cluster (using gcc)

Example:

 $ make atstar

 $ ./bin/pdpjes [--noomp] [--serial] datafiles...

Example:

 $ ./bin/pdpjes --serial data/fixtures/kun01.txt data/fixtures/kun03.txt