To understand the many different phenomena occurring in the nuclear environment, materials studies  needs to encompass broad time and length scales starting from atomistic descriptions of defect energetics and ending with a description of bulk property behaviour at the continuum limit. A single code running on the supercomputers of today or even those available in the future cannot describe all these phenomena. Methods undertaken to understand the mechanical properties of these materials include state of the art multi-scale, multi-code computations and multi-dimensional experiments. These simulations begin at the atomistic level with ab initio, molecular dynamics and kinetic Monte Carlo techniques, moves through the meso-scale using mean field rate theory and Dislocation Dynamics, and end with the macro-scale using Calphad, Finite Element methods and continuum models. The vision is that such a multiscale modelling and experimental approach will probe beyond currently possible approaches to become a predictive tool in estimating mechanical properties and eventually the lifetimes of materials. In the future, it is envisaged that development of tailor-made alloys and ceramics with optimized composition would be possible as an outcome of accurate materials modelling. To realize such a scheme, it is necessary, at an international level, to include modelling as a constitutive part of materials research.