Predictive first-principles simulations for co-designing next-generation energy-efficient AI systems

Abstract

In modern generative-AI workloads, matrix-vector/matrix-matrix multiplications (\emph{MatMul}) dominate the compute and energy cost. Achieving dramatic reductions in energy per token therefore requires a novel, specialized hardware that is co-designed across materials, devices, interconnects, circuits, and architectures rather than optimized at any single layer in isolation. In this \emph{Perspectives} article, we argue that \emph{predictive} (first-principles, fitting-parameter-free) device and interconnect simulations can close the loop between nanoscale physics and workload-level metrics, enabling the identification of device/interconnect operating regimes that plausibly support \emph{orders-of-magnitude} improvements in energy efficiency of AI accelerators.

References in corpus (8)

• Density functional method for nonequilibrium electron transport
• QuantumATK: An integrated platform of electronic and atomic-scale modelling tools
• Improvements on non-equilibrium and transport Green function techniques: the next-generation transiesta
• Efficient self-consistent quantum transport simulator for quantum devices
• Conductivity and size quantization effects in semiconductor $\delta$-layer systems
• Process Variation-Aware Compact Model of Strip Waveguides for Photonic Circuit Simulation
• Uncovering anisotropic effects of electric high-moment dipoles on the tunneling current in $\delta$-layer tunnel junctions
• Influence of imperfections on tunneling rate in $\delta$-layer junctions