Nanos Group, Barcelona Supercomputing Centre
Contact: Alex Duran (firstname.lastname@example.org)
An LU matrix factorization over sparse matrices is computed. A first level matrix is composed by pointers to small submatrices that may not be allocated. Due to the sparseness of the matrix, a lot of imbalance exists. Matrix size and submatrix size can be set at execution time. While a dynamic schedule can reduce the imbalance, a solution with task-based parallelism seems to obtain better results. In each of the sparseLU phases, a task is created for each block of the matrix that is not empty.
The program takes two input size n and m from the command line and generates a sparse matrix with those sizes. The reference data set is 120 and 501, the train data set is 100 and 25, and the test data set is 50 and 25.
For benchmarking purposes, where the SPEC tools are used to validate the solution, the computed results are only stored.
A portion of the solved matrix is produced.
A. Duran, X. Teruel, R. Ferrer, X. Martorell and E. Ayguadé: Barcelona OpenMP Tasks Suite: A Set of Benchmarks Targeting the Exploitation of Task Parallelism in OpenMP , in: Proceedings of ICPP-2009, The 38th International Conference on Parallel Processing, Copyright 2009 The Institute of Electrical and Electronics Engineers, Inc. DOI
Last Updated: March 15, 2012