Due to their unique structural characteristics such as long-range disorder and short-range order, metallic glasses exhibit complex and diverse macroscopic plastic deformation behaviors. In spite of this, the field of amorphous alloy has currently reached a consensus: complex and diverse macroscopic plastic deformation behaviors of metallic glasses have the same microscopic "unit process", that is a local rearrangement of atoms undergoing inelastic shear distortion to accommodate plastic deformation, commonly referred to as "flow event". Since the size of the "flow event" is too small and its occurrence is too fast, the existing testing methods, which are difficult to achieve high time and spatial resolutions at the same time, are hard to directly observe it. So people often adopt simulation methods to carry on the research on the "flow event". In this paper, we briefly review these simulation methods and their advantages and disadvantages. Finally, we identify a number of important points that deserve further investigation in the simulation research on the fundamental unit for the plastic deformation of metallic glass.