ABSTRACT
The explosion of nanophotonic design over the past two decades has led to amazing breakthroughs in areas such as telecommunications, biological and physical sciences, integrated circuits, and computing. Behind this technology lies an integral, often overlooked factor in the design process: simulation. Before the devices of tomorrow are fabricated, they are first modeled and extensively simulated. It is here that new ideas are tested, modified, retested, and constantly manipulated to create the optimal device. Unfortunately, as nanophotonic designs have grown ever more complicated, researchers have taken the current generation of modeling and simulation tools to their limits. Despite the wealth of computational algorithms available to model such devices, current computer system technology remains unable to fully and accurately model many devices and systems. In fact, the lack of a powerful simulation platform has directly impeded progress in this field. For example, in their 2003
Optics Express
paper on negative index materials (NIMs), Greegor et al. state that ‘‘the direct detailed simulation of a large number of unit cells is not possible due to computer memory and computational time requirements” [1]. Although supercomputers and clusters of computers can be used to shorten the computational time, these solutions can
be prohibitively expensive and frequently impractical. As a result, an approach that increases simulation speeds in a relatively inexpensive and practical way is required.
