15 | | - ToDo: |
16 | | ... |
17 | | }}} |
| 10 | == Technologies we used == |
| 11 | |
| 12 | * [https://computing.llnl.gov/tutorials/openMP/] (OPENMP) |
| 13 | * [https://computing.llnl.gov/tutorials/mpi/] (MPI) |
| 14 | * Languages: C/C++ |
| 15 | |
| 16 | == Steps == |
| 17 | |
| 18 | 1.Search and see which graph algorithms can be parallelized |
| 19 | |
| 20 | 2.Implement the serial versions of the graph algorithms we chose |
| 21 | |
| 22 | 3.Build case tests with some hand made generators to verify |
| 23 | the correctness of the serial algorithms |
| 24 | |
| 25 | 4.Implement in OpenMP and MPI the parallel versions of the graph algorithms |
| 26 | |
| 27 | 5.Test the performance and correctness of the parallel algorithms on the fep cluster |
| 28 | |
| 29 | |
| 30 | == Results == |
| 31 | |
| 32 | The results we obtained are in the powerpoint presentation located in the svn repository |
| 33 | in the folder named Presentation. |
| 34 | |
| 35 | == References == |
| 36 | |
| 37 | All the articles and books that we used are on the svn repository in the |
| 38 | Documentation folder.They include articles on distributed algorithms for graphs, |
| 39 | some articles on parallel CREW PRAM algorithms for graphs and the Introduction |
| 40 | to Algorithms(doesn't need any introduction) book which was the main reference |
| 41 | for the serial algorithms. |
| 42 | |
| 43 | |
| 44 | |
| 45 | |
| 46 | |
| 47 | |
| 48 | |
| 49 | |
| 50 | |
| 51 | |
| 52 | |
| 53 | |
| 54 | |
| 55 | |
| 56 | |
| 57 | |
| 58 | |
| 59 | |
| 60 | |
| 61 | |
| 62 | |