ithink20 / pagerank

C implementation of the PageRank algorithm, with and without parallelization.

PageRank

aka this was interesting.

C implementation of the PageRank algorithm, with and without parallelization. Used as a group project for the High Performance Computing course held at Ca' Foscari University of Venice, master's degree in Computer science. The algorithm is implemented sequentially and then parallelized using the openMP library.

The code

Several files are included:

• step1.c, sequential implementation of PageRank. Uses a transposed adjacency matrix;
• step2.c, sequential implementation of PageRank. Uses compressed sparse row organization of the adjacency matrix;
• step3.c, parallel implementation of PageRank. Customize scheduling type (static, dynamic) and number of threads;
• step2mmap.c, same as step2.c but using mmap;
• step3mmap.c, same as step3.c but using mmap.

Compile the code using these data sets and see the magic.

Acknowledgements

This project was done with with Gaia O. and Gianluca C..