Linux software for large topology optimization problems

Recently the commercial finite element tool-box COMSOL has proven to be an excellent platform for development of the topology optimization method, teaching and other multi-physics modeling problems. This is mainly due to COMSOL’s flexibility and symbolic infrastructure which puts focus on the modeling of the physics using partial differential equations (PDEs) rather than the numerical implementation. However, COMSOL lacks computational performance for large scale 3D problems, which are typically solved using an iterative strategy on a multi processor computer. Although COMSOL is a rapidly evolving product, which allows a parallel solution of the PDE, it lacks the important feature that the matrix-generation part of the computations is localized to each processor. This is well-known to be critical for obtaining a useful speedup on a Linux cluster and it motivates the search for a COMSOL-like package for large topology optimization problems. One candidate for such software is developed for Linux by Sandia Nat’l Lab in the USA being the Sundance system. Sundance also uses a symbolic representation of the PDE and a scalable numerical solution is achieved by employing the underlying Trilinos library. This talk investigates the efficiency of Sundance for large topology optimization problems.