University of Virginia
Dense Linear Algebra, Linear Algebra
Gaussian Elimination computes result row by row, solving for all of the variables in a linear system. The algorithm must synchronize between iterations, but the values calculated in each iteration can be computed in parallel.
This benchmark takes the size of the matrix to factor as input on the command line. It uses an internally generated matrix as data.
The solution vector is used as output for validation.
The output file ge.out contains detailed timing information about the run. It also shows which device was selected along with what devices where available to OpenCL.
 S. Che, J. W. Sheaffer, M. Boyer, L. G. Szafaryn, L. Wang, and K. Skadron. A Characterization of the Rodinia Benchmark Suite with Comparison to Contemporary CMP Workloads. In Proceedings of the IEEE International Symposium on Workload Characterization, Dec. 2010.