CUA OpenMP Nonlinear Optimization Tool

Nonlinear programming (NLP) allows for the solution of complex engineering problems, however, none of the currently available solvers capitalizes on parallel computing. Many NASA trajectory design packages (OTIS, EMTG, MALTO) have already had their own code streamlined, and it is now the serial execution of existing NLP solvers that represents the largest bottleneck. CU Aerospace has an existing prototype of this kind of solver, the Nonlinear Parallel Optimization Tool (NLPAROPT), which has already demonstrated speed superiority over comparable serial algorithms and shown that there remains significant potential for improvements. Currently, NLPAROPT is restricted to run on distributed memory systems. It is the goal of this Phase I effort to create a sister program to NLPAROPT, the CUA OpenMP Nonlinear Optimization Tool (COMPNOT), which will be compatible with shared memory systems. As large-scale shared memory parallel systems, such as Intel's Xeon Phi family, become more commercially available, COMPNOT will greatly expand the market for this NLP solver, even enabling most modern desktop computers to effectively run it. Additionally, Phase I will entail developing hardware-specific optimization, focusing on the Intel Math Kernel Library (MKL), but other platforms will be explored as well. At the end of Phase I, can begin integration into NASA trajectory design packages, significantly reducing the time-to-solution.