Abstract | ||
---|---|---|
We describe an implementation of a parallel linear programming solver on the Cell BE processor. This implementation is based on GLPK C routines which solve LP problems using a serial implementation of one of the interior point methods. We have identified the computational kernels of the serial version and decided to implement a parallel version of Cholesky factorization and integrate it into GLPK. Our decision stemmed from the fact that Cholesky factorization is the most computationally expensive kernel that has the most potential to parallelize efficiently on the Cell BE processor. Compared to the execution time of serial GLPK on the Cell Power Processing Element (PPE), we were able to obtain a speedup of up to 7 when solving some of the large size Netlib problems on Sony's PlayStation 3. (C) 2010 Published by Elsevier Ltd. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1016/j.procs.2010.04.117 | Procedia Computer Science |
Keywords | Field | DocType |
Cell BE processor,Linear Programming,Sparse Cholesky Factorization,Interior Point Methods | Kernel (linear algebra),Mathematical optimization,Netlib,Computer science,Parallel computing,Minimum degree algorithm,Linear programming,Solver,Interior point method,Speedup,Cholesky decomposition | Journal |
Volume | Issue | ISSN |
1 | 1 | 1877-0509 |
Citations | PageRank | References |
1 | 0.38 | 9 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mujahed Eleyat | 1 | 7 | 2.98 |
Lasse Natvig | 2 | 109 | 19.61 |