In this paper, a comprehensive survey for the state of art of a posteriori error estimates in finite element computation for magnetostatics since 1980s is presented.
But in terms of motivation since, is the fact that what defines sort of a high-performance computer or supercomputer such a moving target where state of art today is yesterday's news three years from now?
This is a $1 trillion project and they're estimating that it's going to cost even more because all of these technologies still have to be developed and they're really trying to get the state of art for everything.