Parallel SpGEMM
From Cs240aproject
| Line 9: | Line 9: | ||
1) The simplest decomposition is the regular block checkerboard partitioning [http://www.parallab.uib.no/projects/molecul/matrix/main/node12.html]. Here we assume assumes the existence of a fast sequential algorithm for sparse matrix multiplication that is used as a block box. | 1) The simplest decomposition is the regular block checkerboard partitioning [http://www.parallab.uib.no/projects/molecul/matrix/main/node12.html]. Here we assume assumes the existence of a fast sequential algorithm for sparse matrix multiplication that is used as a block box. | ||
| - | 2) The second way to decompose to data is to try assigning equal number of edges to each processor. The initial idea is to map each non-zero of the matrix to a point in the rectangular (m x n) region and use a quadtree data structure. Here one may or may not use a sequential matrix multiplication as block box. | + | |
| + | 2) The second way to decompose to data is to try assigning equal number of edges to each processor. The initial idea is to map each non-zero of the matrix to a point in the rectangular (m x n) region and use a quadtree data structure. Here one may or may not use a sequential matrix multiplication as block box [Yet to be explored]. | ||
| + | |||
| + | 3) The third and the hardest way is to assign edges to processors such that the following metric is minimized. | ||
| + | <math>min{max_i{communication(i) + computation(i)}}</math> | ||
Revision as of 08:47, 9 May 2007
This is the Spring 2007 project of Aydin Buluc and Fenglin Liao.
Project Description
The project is to write a 2D parallel implementation of the sparse matrix times sparse matrix routine. It is projected to be implemented in C++ using MPI. The project will have multiple iterations:
1) The simplest decomposition is the regular block checkerboard partitioning [1]. Here we assume assumes the existence of a fast sequential algorithm for sparse matrix multiplication that is used as a block box.
2) The second way to decompose to data is to try assigning equal number of edges to each processor. The initial idea is to map each non-zero of the matrix to a point in the rectangular (m x n) region and use a quadtree data structure. Here one may or may not use a sequential matrix multiplication as block box [Yet to be explored].
3) The third and the hardest way is to assign edges to processors such that the following metric is minimized.
Failed to parse (Can't write to or create math temp directory): min{max_i{communication(i) + computation(i)}}
