GPU computations for finite element optimization: A review of the methodology and problems for study

In finite element optimization, the computational load is a limiting issue. Parallelization has been the preferred route to overcome this problem but was again limited by the cost of computers and the number of processors available. The graphics processing unit (GPU) on a PC provides a means of implementing the massive computations on numerous parallel threads cheaply on PCs. This paper lays out recent work which is the way forward for making finite element optimization routine in the design of power frequency devices and nondestructive evaluation (NDE). Going beyond a review we present problems to be examined in GPU computations as fodder for research.