Parallelization strategies for the UTD codes

The structure of the ray-theoretic formulation such as the UTD, shows that in principle its parallelization can be implemented on parallel computers both with high and low number of processing elements. A significant reduction in execution time can be achieved even with a top down scheme, starting with the shaping parameter on a moderately parallel machine, as implemented on the Flosolver at NAL. It is also observed that efficiency would further improve for the UTD codes if the parallelization is done at the level of the first geodesic constant determination of the surface geodesics.